# ===== Imports =====

# ===== General =====
import numpy as np
import pandas as pd
import math
import warnings
warnings.filterwarnings('ignore')

# ===== Visualization =====
import seaborn as sns
import missingno as msno
import matplotlib.pyplot as plt
import scipy.stats as stats
from matplotlib import patheffects
from matplotlib.colors import LinearSegmentedColormap
import matplotlib.patches as mpatches
import matplotlib.colors as mcolors
import matplotlib.patheffects as path_effects
%matplotlib inline

# ===== Hypotheses testing =====
from scipy.stats import chi2_contingency

# ===== Preprocessing =====
from sklearn.preprocessing import StandardScaler, LabelEncoder, OneHotEncoder, OrdinalEncoder
from sklearn.impute import SimpleImputer
from sklearn.compose import ColumnTransformer
from sklearn.pipeline import Pipeline

# ===== Outlier Influence =====
from statsmodels.stats.outliers_influence import variance_inflation_factor
from statsmodels.tools.tools import add_constant
import statsmodels.api as sm

# ===== Imbalanced handling =====
from imblearn.over_sampling import SMOTE
from sklearn.preprocessing import label_binarize

# ===== Model Selection =====
import time
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LogisticRegression
from sklearn.ensemble import RandomForestClassifier
from xgboost import XGBClassifier
from lightgbm import LGBMClassifier
from sklearn.model_selection import GridSearchCV
from sklearn.model_selection import RandomizedSearchCV, cross_val_score

# ===== Evaluation Metrics =====
from sklearn.metrics import (accuracy_score, precision_score, recall_score, f1_score, roc_auc_score, confusion_matrix, roc_curve, precision_recall_curve, classification_report)
from sklearn.metrics import (precision_recall_curve, roc_curve, auc, confusion_matrix, ConfusionMatrixDisplay)
from sklearn.calibration import calibration_curve
from sklearn.model_selection import StratifiedKFold, cross_val_score

# ===== Load Data =====
df = pd.read_csv('dataset_35_dermatology.csv')

# ===== Checking first five rows of dataset =====
df.head(5).T

# ===== Checking last five rows of dataset =====
df.tail(5).T

# ===== Basic Overview =====

# ===== To view the summary stats of numerical columns =====
df.describe().T

# ===== To View the categorical columns =====
df.describe(include='O').T

# ===== Checking the info of dataset =====
df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 366 entries, 0 to 365
Data columns (total 35 columns):
 #   Column                                    Non-Null Count  Dtype 
---  ------                                    --------------  ----- 
 0   erythema                                  366 non-null    int64 
 1   scaling                                   366 non-null    int64 
 2   definite_borders                          366 non-null    int64 
 3   itching                                   366 non-null    int64 
 4   koebner_phenomenon                        366 non-null    int64 
 5   polygonal_papules                         366 non-null    int64 
 6   follicular_papules                        366 non-null    int64 
 7   oral_mucosal_involvement                  366 non-null    int64 
 8   knee_and_elbow_involvement                366 non-null    int64 
 9   scalp_involvement                         366 non-null    int64 
 10  family_history                            366 non-null    int64 
 11  melanin_incontinence                      366 non-null    int64 
 12  eosinophils_in_the_infiltrate             366 non-null    int64 
 13  PNL_infiltrate                            366 non-null    int64 
 14  fibrosis_of_the_papillary_dermis          366 non-null    int64 
 15  exocytosis                                366 non-null    int64 
 16  acanthosis                                366 non-null    int64 
 17  hyperkeratosis                            366 non-null    int64 
 18  parakeratosis                             366 non-null    int64 
 19  clubbing_of_the_rete_ridges               366 non-null    int64 
 20  elongation_of_the_rete_ridges             366 non-null    int64 
 21  thinning_of_the_suprapapillary_epidermis  366 non-null    int64 
 22  spongiform_pustule                        366 non-null    int64 
 23  munro_microabcess                         366 non-null    int64 
 24  focal_hypergranulosis                     366 non-null    int64 
 25  disappearance_of_the_granular_layer       366 non-null    int64 
 26  vacuolisation_and_damage_of_basal_layer   366 non-null    int64 
 27  spongiosis                                366 non-null    int64 
 28  saw-tooth_appearance_of_retes             366 non-null    int64 
 29  follicular_horn_plug                      366 non-null    int64 
 30  perifollicular_parakeratosis              366 non-null    int64 
 31  inflammatory_monoluclear_inflitrate       366 non-null    int64 
 32  band-like_infiltrate                      366 non-null    int64 
 33  Age                                       366 non-null    object
 34  class                                     366 non-null    int64 
dtypes: int64(34), object(1)
memory usage: 100.2+ KB

# ===== Checking the no. of rows and columns =====
df.shape

(366, 35)

# ===== Domain Analysis =====
df.columns

Index(['erythema', 'scaling', 'definite_borders', 'itching',
       'koebner_phenomenon', 'polygonal_papules', 'follicular_papules',
       'oral_mucosal_involvement', 'knee_and_elbow_involvement',
       'scalp_involvement', 'family_history', 'melanin_incontinence',
       'eosinophils_in_the_infiltrate', 'PNL_infiltrate',
       'fibrosis_of_the_papillary_dermis', 'exocytosis', 'acanthosis',
       'hyperkeratosis', 'parakeratosis', 'clubbing_of_the_rete_ridges',
       'elongation_of_the_rete_ridges',
       'thinning_of_the_suprapapillary_epidermis', 'spongiform_pustule',
       'munro_microabcess', 'focal_hypergranulosis',
       'disappearance_of_the_granular_layer',
       'vacuolisation_and_damage_of_basal_layer', 'spongiosis',
       'saw-tooth_appearance_of_retes', 'follicular_horn_plug',
       'perifollicular_parakeratosis', 'inflammatory_monoluclear_inflitrate',
       'band-like_infiltrate', 'Age', 'class'],
      dtype='object')

# ===== Convert 'Age' column from object to numeric =====
df['Age'] = pd.to_numeric(df['Age'], errors='coerce')
df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 366 entries, 0 to 365
Data columns (total 35 columns):
 #   Column                                    Non-Null Count  Dtype  
---  ------                                    --------------  -----  
 0   erythema                                  366 non-null    int64  
 1   scaling                                   366 non-null    int64  
 2   definite_borders                          366 non-null    int64  
 3   itching                                   366 non-null    int64  
 4   koebner_phenomenon                        366 non-null    int64  
 5   polygonal_papules                         366 non-null    int64  
 6   follicular_papules                        366 non-null    int64  
 7   oral_mucosal_involvement                  366 non-null    int64  
 8   knee_and_elbow_involvement                366 non-null    int64  
 9   scalp_involvement                         366 non-null    int64  
 10  family_history                            366 non-null    int64  
 11  melanin_incontinence                      366 non-null    int64  
 12  eosinophils_in_the_infiltrate             366 non-null    int64  
 13  PNL_infiltrate                            366 non-null    int64  
 14  fibrosis_of_the_papillary_dermis          366 non-null    int64  
 15  exocytosis                                366 non-null    int64  
 16  acanthosis                                366 non-null    int64  
 17  hyperkeratosis                            366 non-null    int64  
 18  parakeratosis                             366 non-null    int64  
 19  clubbing_of_the_rete_ridges               366 non-null    int64  
 20  elongation_of_the_rete_ridges             366 non-null    int64  
 21  thinning_of_the_suprapapillary_epidermis  366 non-null    int64  
 22  spongiform_pustule                        366 non-null    int64  
 23  munro_microabcess                         366 non-null    int64  
 24  focal_hypergranulosis                     366 non-null    int64  
 25  disappearance_of_the_granular_layer       366 non-null    int64  
 26  vacuolisation_and_damage_of_basal_layer   366 non-null    int64  
 27  spongiosis                                366 non-null    int64  
 28  saw-tooth_appearance_of_retes             366 non-null    int64  
 29  follicular_horn_plug                      366 non-null    int64  
 30  perifollicular_parakeratosis              366 non-null    int64  
 31  inflammatory_monoluclear_inflitrate       366 non-null    int64  
 32  band-like_infiltrate                      366 non-null    int64  
 33  Age                                       358 non-null    float64
 34  class                                     366 non-null    int64  
dtypes: float64(1), int64(34)
memory usage: 100.2 KB

# ===== Extracting categorical and numerical columns =====
num_col = [col for col in df.columns if df[col].dtype != 'object']

# ===== Looking at unique values in categorical and numerical columns =====
print("\nNumerical Columns:\n")
for col in num_col:
    print(f'\n{col}:\n{df[col].unique()}')

Numerical Columns:


erythema:
[2 3 1 0]

scaling:
[2 3 1 0]

definite_borders:
[0 3 2 1]

itching:
[3 2 0 1]

koebner_phenomenon:
[0 1 2 3]

polygonal_papules:
[0 3 2 1]

follicular_papules:
[0 3 1 2]

oral_mucosal_involvement:
[0 3 2 1]

knee_and_elbow_involvement:
[1 0 3 2]

scalp_involvement:
[0 1 2 3]

family_history:
[0 1]

melanin_incontinence:
[0 1 2 3]

eosinophils_in_the_infiltrate:
[0 2 1]

PNL_infiltrate:
[0 1 3 2]

fibrosis_of_the_papillary_dermis:
[0 3 1 2]

exocytosis:
[3 1 0 2]

acanthosis:
[2 3 1 0]

hyperkeratosis:
[0 2 1 3]

parakeratosis:
[0 2 3 1]

clubbing_of_the_rete_ridges:
[0 2 1 3]

elongation_of_the_rete_ridges:
[0 2 3 1]

thinning_of_the_suprapapillary_epidermis:
[0 2 3 1]

spongiform_pustule:
[0 2 1 3]

munro_microabcess:
[0 1 2 3]

focal_hypergranulosis:
[0 2 3 1]

disappearance_of_the_granular_layer:
[0 3 2 1]

vacuolisation_and_damage_of_basal_layer:
[0 2 3 1]

spongiosis:
[3 0 2 1]

saw-tooth_appearance_of_retes:
[0 2 3 1]

follicular_horn_plug:
[0 1 2 3]

perifollicular_parakeratosis:
[0 2 1 3]

inflammatory_monoluclear_inflitrate:
[1 2 3 0]

band-like_infiltrate:
[0 3 1 2]

Age:
[55.  8. 26. 40. 45. 41. 18. 57. 22. 30. 20. 21. 10. 65. 38. 23. 17. 51.
 42. 44. 33. 43. 50. 34. nan 15. 46. 62. 35. 48. 12. 52. 60. 32. 19. 29.
 25. 36. 13. 27. 31. 28. 64. 39. 47. 16.  0.  7. 70. 37. 61. 67. 56. 53.
 24. 58. 49. 63. 68.  9. 75.]

class:
[2 1 3 5 4 6]

# ===== Fill NaN with median =====
df['Age'] = df['Age'].fillna(df['Age'].median())

# ===== Finally cast to integer =====
df['Age'] = df['Age'].astype(int)

# ===== Check duplicate values =====
# ===== Total number of rows =====
total_rows = len(df)

# ===== Count duplicate rows =====
duplicate_count = df.duplicated().sum()

# ===== Percentage of duplicates =====
duplicate_percentage = (duplicate_count / total_rows) * 100

print(f"Total Rows: {total_rows}")
print(f"Duplicate Rows: {duplicate_count}")
print(f"Percentage of Duplicates: {duplicate_percentage:.2f}%")

Total Rows: 366
Duplicate Rows: 0
Percentage of Duplicates: 0.00%

# ===== List of clinical features related to a medical condition. =====
clinical_features = [
    'erythema',
    'scaling',
    'definite_borders',
    'itching',
    'koebner_phenomenon',
    'polygonal_papules',
    'follicular_papules',
    'oral_mucosal_involvement',
    'knee_and_elbow_involvement',
    'scalp_involvement',
    'family_history'
]

# ===== Clinical Features =====
# ====== Create Grid ======
n_features = len(clinical_features)
rows = (n_features // 3) + int(n_features % 3 != 0)
plt.figure(figsize=(24, 16))

for i, feature in enumerate(clinical_features, 1):
    ax = plt.subplot(rows, 3, i)
    sns.countplot(x=feature, data=df, palette=["red", "navy"], ax=ax)
    ax.set_title(feature.replace("_", " ").title(), fontsize=12, fontweight="bold")
    ax.set_xlabel("")
    ax.set_ylabel("Count")
    ax.set_facecolor("lightgray")

    for p in ax.patches:
        ax.annotate(
            f'{int(p.get_height())}',
            (p.get_x() + p.get_width() / 2., p.get_height()),
            ha='center', va='bottom', fontsize=10, fontweight="bold", color="black"
        )

plt.suptitle("Distribution of Clinical Features", fontsize=18, fontweight="bold", y=1.02)
plt.tight_layout()
plt.show()

# ===== Age Feature =====
plt.figure(figsize=(23,7))
ax = sns.countplot(x="Age", data=df, palette=["red", "navy"])
ax.set_title("Distribution of Age Feature", fontsize=14, fontweight="bold")
ax.set_xlabel("Age")
ax.set_ylabel("Count")
plt.xticks(rotation=90)
ax.set_facecolor("lightgray")
plt.show()

# ===== Histopathological Feature =====
# ===== List of histopathological features related to a medical condition =====

Histopath_Features = [
    'melanin_incontinence',
    'eosinophils_in_the_infiltrate',
    'PNL_infiltrate',
    'fibrosis_of_the_papillary_dermis',
    'exocytosis', 'acanthosis',
    'hyperkeratosis',
    'parakeratosis',
    'clubbing_of_the_rete_ridges',
    'elongation_of_the_rete_ridges',
    'thinning_of_the_suprapapillary_epidermis',
    'spongiform_pustule',
    'munro_microabcess',
    'focal_hypergranulosis',
    'disappearance_of_the_granular_layer',
    'vacuolisation_and_damage_of_basal_layer',
    'spongiosis',
    'saw-tooth_appearance_of_retes',
    'follicular_horn_plug',
    'perifollicular_parakeratosis',
    'inflammatory_monoluclear_inflitrate',
    'band-like_infiltrate'
]

# ===== Histopathological Feature Run Code =====
n_cols = 5
n_rows = math.ceil(len(Histopath_Features) / n_cols)
fig, axes = plt.subplots(n_rows, n_cols, figsize=(25, 22))
axes = axes.flatten()
for i, feature in enumerate(Histopath_Features):
    ax = axes[i]
    sns.countplot(
        x=df[feature],
        ax=ax,
        palette=["red", "navy"]
    )
    ax.set_title(feature, fontsize=13, fontweight="bold")
    ax.set_facecolor("lightgray")
    ax.set_xlabel("")
    ax.set_ylabel("")
for j in range(i+1, len(axes)):
    fig.delaxes(axes[j])

fig.suptitle("Histopathological Features", fontsize=18, fontweight="bold")
plt.tight_layout(rect=[0, 0, 1, 0.97])
plt.show()

# ===== Class Feature =====
df_class = df.copy()
df_class['class'].replace(
    [1,2,3,4,5,6],
    ['Psoriasis', 'Seboreic_Dermatitis', 'Lichen_Planus',
     'Pityriasis_Rosea','Cronic_Dermatitis','Pityriasis_rubra_pilaris'],
    inplace=True
)

# ===== Run code =====
class_counts = df_class['class'].value_counts().reset_index()
class_counts.columns = ['class', 'count']
plt.figure(figsize=(20,7))
ax = sns.barplot(
    data=class_counts,
    x="class",
    y="count",
    palette=sns.color_palette(["red", "navy"]),
)
ax.set_facecolor("lightgray")

plt.title("Distribution of Dermatological Classes", fontsize=18, fontweight="bold", pad=15)
plt.xlabel("Disease Class", fontsize=14, fontweight="bold")
plt.ylabel("Number of Cases", fontsize=14, fontweight="bold")
plt.xticks(rotation=25, ha="right", fontsize=12, fontweight="medium")
plt.yticks(fontsize=12)

for p in ax.patches:
    ax.annotate(
        f"{int(p.get_height())}",
        (p.get_x() + p.get_width() / 2., p.get_height()),
        ha='center', va='bottom',
        fontsize=11, fontweight="bold", color="black"
    )

sns.despine(left=True, bottom=True)
plt.tight_layout()
plt.show()

# ===== Run code =====
class_counts = df_class['class'].value_counts().reset_index()
class_counts.columns = ['class', 'count']
colors = ["red", "navy", "green", "orange", "purple", "teal"]

plt.figure(figsize=(20,5))
wedges, texts, autotexts = plt.pie(
    class_counts['count'],
    labels=class_counts['class'],
    autopct='%1.1f%%',
    startangle=140,
    colors=colors,
    wedgeprops={'edgecolor': 'black'},
    textprops={'fontsize': 12, 'weight': 'bold'}
)
centre_circle = plt.Circle((0,0), 0.70, fc='lightgray')
fig = plt.gcf()
fig.gca().add_artist(centre_circle)
plt.title("Distribution of Dermatological Classes", fontsize=18, fontweight="bold", pad=15)
plt.tight_layout()
plt.show()

# ===== Run code =====
plt.figure(figsize=(18,6))
ax = sns.histplot(
    df['Age'],
    bins=20,
    kde=True,
    color="navy",
    edgecolor="black",
    alpha=0.8
)
plt.hist(df['Age'], bins=20, color="navy", alpha=0.3, edgecolor="black")

# ===== Background =====
ax.set_facecolor("lightgray")

# ===== Mean & Median lines =====
mean_age = df['Age'].mean()
median_age = df['Age'].median()
plt.axvline(mean_age, color="red", linestyle="--", linewidth=2, label=f"Mean = {mean_age:.1f}")
plt.axvline(median_age, color="green", linestyle="-.", linewidth=2, label=f"Median = {median_age:.1f}")

plt.title("Age Distribution", fontsize=18, fontweight="bold", pad=15)
plt.xlabel("Age", fontsize=14, fontweight="bold")
plt.ylabel("Frequency", fontsize=14, fontweight="bold")
plt.legend(fontsize=12, frameon=True, facecolor="white")
plt.tight_layout()
plt.show()

# ===== Regression plot of Age feature vs Target Variable =====
df_class['class'] = df_class['class'].astype('category')
class_mapping = dict(zip(df_class['class'].cat.categories, df_class['class'].cat.codes))

colors = ["red", "navy", "green", "orange", "purple", "teal"]
palette = sns.color_palette(colors, n_colors=df_class['class'].nunique())
class_colors = dict(zip(df_class['class'].cat.categories, palette))

# ===== Create figure =====
plt.figure(figsize=(17,6))

# ===== Scatterplot colored by class =====
ax = sns.scatterplot(
    data=df_class,
    x="Age",
    y="class",
    hue="class",
    palette=class_colors,
    s=80,
    alpha=0.7,
    edgecolor="black"
)

for cls, color in class_colors.items():
    subset = df_class[df_class['class'] == cls]
    sns.regplot(
        data=subset,
        x="Age",
        y=subset['class'].map(class_mapping),
        scatter=False,
        line_kws={"linewidth":2, "color":color}
    )

ax.set_ylabel("")
ax.set_yticks([])
ax.set_facecolor("lightgray")
plt.title("Relationship Between Age Feature and Target Variable(Class)", fontsize=18, fontweight="bold", pad=15)
plt.xlabel("Age", fontsize=14, fontweight="bold")
plt.legend(title="Disease Class", fontsize=11, title_fontsize=12, frameon=True, facecolor="white")
plt.tight_layout()
plt.show()

# ===== All Features vs Target Variable =====
# ===== Drop 'Age' from df_class =====
numeric = df_class.drop(columns=['Age'], errors='ignore')
sns.set_style("whitegrid")

# ===== Custom class colors =====
colors = ["red", "navy", "green", "orange", "purple", "teal"]
unique_classes = numeric['class'].unique()
class_palette = dict(zip(unique_classes, colors[:len(unique_classes)]))

# ===== Figure layout =====
n_cols = 3
n_rows = int(np.ceil((numeric.shape[1] - 1) / n_cols))
plt.figure(figsize=(28, n_rows * 5))
plotnumber = 1

for column in numeric.columns:

    if column == 'class':
        continue

    ax = plt.subplot(n_rows, n_cols, plotnumber)

    ax.set_facecolor("lightgray")

    # ===== Countplot with custom colors =====
    sns.countplot(x=column, data=numeric, hue='class', palette=class_palette, edgecolor='black', ax=ax)

    # ===== Add counts on top of bars =====
    for p in ax.patches:
        height = p.get_height()
        if height > 0:
            ax.text(p.get_x() + p.get_width()/2.,
                    height + 0.05*height,
                    '{:1.0f}'.format(height),
                    ha="center", fontsize=10, color='black', fontweight='bold')

    ax.set_xticklabels(ax.get_xticklabels(), rotation=45, ha='right', fontsize=12)
    ax.set_xlabel(column.title(), fontsize=12, fontweight='bold')
    ax.set_ylabel('Count', fontsize=12, fontweight='bold')
    ax.set_title(f'{column.title()} Distribution by Class', fontsize=14, fontweight='bold', pad=10)
    sns.despine(ax=ax)
    plotnumber += 1

plt.tight_layout()
plt.suptitle('Feature Distributions Across Classes', fontsize=20, fontweight='bold', y=1.01)
plt.show()

# ===== Age Feature vs Target Variable =====
if 'Age' in df_class.columns:
    sns.set_style("whitegrid")

    # ===== Custom class colors =====
    colors = ["red", "navy", "green", "orange", "purple", "teal"]
    unique_classes = df_class['class'].unique()
    class_palette = dict(zip(unique_classes, colors[:len(unique_classes)]))

    # ===== Bin Age into intervals =====
    bins = np.arange(df_class['Age'].min(), df_class['Age'].max() + 5, 5)
    df_class['Age_bin'] = pd.cut(df_class['Age'], bins=bins, right=False)

    # ===== Create figure =====
    plt.figure(figsize=(27, 10))
    ax = sns.countplot(
        x='Age_bin',
        data=df_class,
        hue='class',
        palette=class_palette,
        edgecolor='black',
        alpha=0.85
    )

    ax.set_facecolor("lightgray")
    for p in ax.patches:
        height = p.get_height()
        if height > 0:
            ax.text(
                p.get_x() + p.get_width()/2.,
                height + 0.5,
                '{:1.0f}'.format(height),
                ha="center",
                fontsize=10,
                color='black',
                fontweight='bold'
            )

    ax.set_xlabel('Age (Binned in 5-year intervals)', fontsize=14, fontweight='bold')
    ax.set_ylabel('Count', fontsize=14, fontweight='bold')
    ax.set_title('Age Distribution Across Dermatological Classes', fontsize=20, fontweight='bold', pad=20)
    ax.set_xticklabels(ax.get_xticklabels(), ha='right', fontsize=12)
    ax.legend(title='Class', bbox_to_anchor=(1.02, 1), loc='upper left', fontsize=12, title_fontsize=13, frameon=True, facecolor='white')
    sns.despine(ax=ax)
    plt.tight_layout()
    plt.show()

# ===== Correlation Heatmap visualization code =====
numeric_df = df.select_dtypes(include=['number']).copy()

target_col = 'class'
if df[target_col].dtype.name == 'category':
    numeric_df[target_col] = df[target_col].cat.codes

corr_with_target = numeric_df.corr()[target_col].drop(target_col)
top_features = corr_with_target.abs().sort_values(ascending=False).head(15)
top_feature_list = top_features.index.tolist()
corr_top = numeric_df[top_feature_list + [target_col]].corr()

plt.figure(figsize=(30, 13))
custom_cmap = sns.color_palette("blend:blue,white,green", as_cmap=True)
sns.heatmap(
    corr_top,
    annot=True,
    fmt=".2f",
    cmap=custom_cmap,
    center=0,
    linewidths=1.2,
    linecolor="lightgrey",
    annot_kws={"size":12, "weight":"bold", "color":"black"},
    cbar_kws={"shrink":0.7, "aspect":30, "label":"Correlation Strength"}
)
plt.title("Correlated Features with Target", fontsize=18, fontweight="bold", pad=20)
plt.xticks(rotation=45, ha="right", fontsize=11, weight="bold")
plt.yticks(rotation=0, fontsize=11, weight="bold")
plt.tight_layout()
plt.show()

# ===== Create contingency table =====
contingency_erythema = pd.crosstab(df_class['erythema'], df_class['class'])

chi2, p, dof, expected = chi2_contingency(contingency_erythema)
print("Chi-Square Test: Erythema vs Class")
print("Chi2 Statistic:", chi2)
print("Degrees of Freedom:", dof)
print("p-value:", p)

# ===== Interpretation =====
if p < 0.05:
    print("\nResult: Reject H0 → Erythema significantly affects Skin Disorder Class.")
else:
    print("\nResult: Fail to reject H0 → No significant effect of Erythema on Skin Disorder Class.")

Chi-Square Test: Erythema vs Class
Chi2 Statistic: 78.71170847325529
Degrees of Freedom: 15
p-value: 1.200485187555973e-10

Result: Reject H0 → Erythema significantly affects Skin Disorder Class.

# ===== Create contingency table =====
contingency_scaling = pd.crosstab(df_class['scaling'], df_class['class'])
chi2, p, dof, expected = chi2_contingency(contingency_scaling)

print("Chi-Square Test: Scaling vs Class")
print("Chi2 Statistic:", chi2)
print("Degrees of Freedom:", dof)
print("p-value:", p)

# ===== Interpretation =====
if p < 0.05:
    print("\nResult: Reject H0 → Scaling significantly affects Skin Disorder Class.")
else:
    print("\nResult: Fail to reject H0 → No significant effect of Scaling on Skin Disorder Class.")

Chi-Square Test: Scaling vs Class
Chi2 Statistic: 132.09466055613035
Degrees of Freedom: 15
p-value: 8.263440529720813e-21

Result: Reject H0 → Scaling significantly affects Skin Disorder Class.

# ===== Create contingency table =====
contingency_fh = pd.crosstab(df_class['family_history'], df_class['class'])
chi2, p, dof, expected = chi2_contingency(contingency_fh)

print("Chi-Square Test: Family History vs Class")
print("Chi2 Statistic:", chi2)
print("Degrees of Freedom:", dof)
print("p-value:", p)

# ===== Interpretation =====
if p < 0.05:
    print("\nResult: Reject H0 → Family history significantly affects Skin Disorder Class.")
else:
    print("\nResult: Fail to reject H0 → No significant effect of Family history on Skin Disorder Class.")

Chi-Square Test: Family History vs Class
Chi2 Statistic: 77.5603988742236
Degrees of Freedom: 5
p-value: 2.716062051379337e-15

Result: Reject H0 → Family history significantly affects Skin Disorder Class.

# ===== Finding a missing values =====
df.isnull().sum().to_frame("Missing_Values")

# ===== Plotting box plots for all numerical variable =====
age_df = df[['Age']]

# ===== Background style =====
plt.style.use('ggplot')

plt.figure(figsize=(8, 5))
sns.boxplot(
    data=age_df,
    x='Age',
    boxprops=dict(color='blue', facecolor='lightblue', linewidth=2),
    flierprops=dict(marker='o', markerfacecolor='red', markersize=5, linestyle='none'),
    medianprops=dict(color='black', linewidth=2),
    whiskerprops=dict(color='blue', linewidth=2),
    capprops=dict(color='blue', linewidth=2)
)

plt.title("Outlier Visualization: Age", fontsize=18, fontweight='bold')
plt.xlabel("Age", fontsize=14, fontweight='bold')
plt.tight_layout()
plt.show()

# ===== Defining the function for outlier detection and percentage calculation using IQR =====

def detect_outliers(data):
    data = np.array(data)

    # ===== Quartiles =====
    q1 = np.percentile(data, 25)
    q2 = np.percentile(data, 50)
    q3 = np.percentile(data, 75)

    # ===== IQR & boundsa =====
    IQR = q3 - q1
    lower_bound = q1 - 1.5 * IQR
    upper_bound = q3 + 1.5 * IQR

    # ===== Outlier detection =====
    outliers = data[(data < lower_bound) | (data > upper_bound)]
    outlier_count = len(outliers)
    outlier_percent = round(outlier_count * 100 / len(data), 2)

    # ===== Display results =====
    print(f"Q1 = {q1}, Q2 (Median) = {q2:.2f}, Q3 = {q3}")
    print(f"IQR = {IQR:.2f}")
    print(f"Lower Bound = {lower_bound:.2f}, Upper Bound = {upper_bound:.2f}")
    print(f"Outliers Detected: {outlier_count}")
    print(f"Outlier Percentage: {outlier_percent}%\n")

# ===== Calculating IQR, Lower/Upper Bounds, and Outlier Counts for Continuous Numerical Features =====

outliers_age = detect_outliers(df['Age'])
print("Detected outliers in Age column:")
print(outliers_age)

Q1 = 25.0, Q2 (Median) = 35.00, Q3 = 48.0
IQR = 23.00
Lower Bound = -9.50, Upper Bound = 82.50
Outliers Detected: 0
Outlier Percentage: 0.0%

Detected outliers in Age column:
None

# ===== Check if my target feature is imbalanced =====
df['class'].value_counts(normalize=True) * 100

# ===== Visualization code =====
counts = df['class'].value_counts()
labels = counts.index
colors = ["red", "navy", "green", "orange", "purple", "teal"]
explode = [0.05] * len(counts)

plt.figure(figsize=(8,6))
plt.pie(
    counts,
    labels=labels,
    autopct='%1.1f%%',
    startangle=90,
    colors=colors,
    explode=explode,
    shadow=True
)
plt.title("Target Variable Distribution (Class)", fontsize=16, fontweight='bold')
plt.show()

# ===== Select your features wisely to avoid overfitting =====
# ===== Correlation Heatmap visualization code =====

corr = df.corr(numeric_only=True)

custom_cmap = sns.color_palette("blend:blue,white,green", as_cmap=True)

plt.figure(figsize=(30,30))
sns.heatmap(
    corr,
    annot=True,
    fmt=".2f",
    cmap=custom_cmap,
    center=0,
    linewidths=1.5,
    linecolor="lightgrey",
    annot_kws={"size":12, "weight":"bold", "color":"black"},
    cbar_kws={"shrink":0.7, "aspect":30, "label":"Correlation Strength"}
)

plt.title("Top Feature Correlations",
          fontsize=16, fontweight="bold", color="#004d40", pad=20)
plt.xticks(rotation=45, ha="right", fontsize=11, weight="bold", color="#222")
plt.yticks(rotation=0, fontsize=11, weight="bold", color="#222")
plt.grid(False)
plt.tight_layout()
plt.show()

# ===== Defining a function for variance_inflation_factor =====

def calc_vif(df):
    """
    Calculates Variance Inflation Factor (VIF) for each numerical feature in the dataframe.

    Parameters:
        df (pd.DataFrame): Input dataframe with features

    Returns:
        pd.DataFrame: VIF values sorted in descending order
    """
    # ===== Select only numeric columns =====
    X = df.select_dtypes(include=[np.number])

    # ===== Add constant to the model for intercept =====
    X = add_constant(X)

    # ===== Compute VIF for each feature =====
    vif_data = pd.DataFrame()
    vif_data["Feature"] = X.columns
    vif_data["VIF"] = [variance_inflation_factor(X.values, i) for i in range(X.shape[1])]

    # ===== Drop the constant term and sort results =====
    vif_data = vif_data[vif_data["Feature"] != "const"]
    return vif_data.sort_values(by="VIF", ascending=False).reset_index(drop=True)

# ===== Run code =====
df_encoded_vif = df.drop("class", axis=1).copy()
vif_result = calc_vif(df_encoded_vif)
print(vif_result)

                                     Feature        VIF
0    vacuolisation_and_damage_of_basal_layer  17.234025
1                       band-like_infiltrate  13.527991
2                       melanin_incontinence  11.344378
3               perifollicular_parakeratosis  11.064677
4              saw-tooth_appearance_of_retes  10.664021
5                          polygonal_papules   8.516422
6                      focal_hypergranulosis   8.221431
7                clubbing_of_the_rete_ridges   7.982584
8                       follicular_horn_plug   7.918923
9                   oral_mucosal_involvement   7.046405
10  thinning_of_the_suprapapillary_epidermis   6.815498
11             elongation_of_the_rete_ridges   6.136761
12                        follicular_papules   4.109109
13          fibrosis_of_the_papillary_dermis   3.393069
14                knee_and_elbow_involvement   3.352390
15                         scalp_involvement   3.204027
16                                exocytosis   2.904057
17                         munro_microabcess   2.526066
18                                spongiosis   2.464172
19                          definite_borders   2.059011
20                        spongiform_pustule   1.934889
21                            PNL_infiltrate   1.830142
22       disappearance_of_the_granular_layer   1.825448
23                             parakeratosis   1.768826
24                                   scaling   1.711604
25                                   itching   1.644960
26                        koebner_phenomenon   1.610212
27                                acanthosis   1.492661
28                                  erythema   1.470500
29                            family_history   1.301650
30                            hyperkeratosis   1.298062
31                                       Age   1.253764
32       inflammatory_monoluclear_inflitrate   1.250822
33             eosinophils_in_the_infiltrate   1.198826

# ===== Checking =====
df.columns

Index(['erythema', 'scaling', 'definite_borders', 'itching',
       'koebner_phenomenon', 'polygonal_papules', 'follicular_papules',
       'oral_mucosal_involvement', 'knee_and_elbow_involvement',
       'scalp_involvement', 'family_history', 'melanin_incontinence',
       'eosinophils_in_the_infiltrate', 'PNL_infiltrate',
       'fibrosis_of_the_papillary_dermis', 'exocytosis', 'acanthosis',
       'hyperkeratosis', 'parakeratosis', 'clubbing_of_the_rete_ridges',
       'elongation_of_the_rete_ridges',
       'thinning_of_the_suprapapillary_epidermis', 'spongiform_pustule',
       'munro_microabcess', 'focal_hypergranulosis',
       'disappearance_of_the_granular_layer',
       'vacuolisation_and_damage_of_basal_layer', 'spongiosis',
       'saw-tooth_appearance_of_retes', 'follicular_horn_plug',
       'perifollicular_parakeratosis', 'inflammatory_monoluclear_inflitrate',
       'band-like_infiltrate', 'Age', 'class'],
      dtype='object')

# ===== Creating final dataframe considering above selected features =====
final_df= df[['exocytosis', 'munro_microabcess', 'spongiosis', 'definite_borders',
              'spongiform_pustule', 'PNL_infiltrate', 'disappearance_of_the_granular_layer',
              'parakeratosis', 'scaling', 'itching', 'koebner_phenomenon', 'acanthosis',
              'erythema', 'family_history', 'hyperkeratosis', 'Age',
              'inflammatory_monoluclear_inflitrate', 'eosinophils_in_the_infiltrate', 'class']]

# ===== Check a final dataset =====

final_df.head().T

# ===== checking which of the variables are continous in nature =====

for i in final_df.columns:
  print(f"The number of unique counts in feature {i} is: {final_df[i].nunique()}")

The number of unique counts in feature exocytosis is: 4
The number of unique counts in feature munro_microabcess is: 4
The number of unique counts in feature spongiosis is: 4
The number of unique counts in feature definite_borders is: 4
The number of unique counts in feature spongiform_pustule is: 4
The number of unique counts in feature PNL_infiltrate is: 4
The number of unique counts in feature disappearance_of_the_granular_layer is: 4
The number of unique counts in feature parakeratosis is: 4
The number of unique counts in feature scaling is: 4
The number of unique counts in feature itching is: 4
The number of unique counts in feature koebner_phenomenon is: 4
The number of unique counts in feature acanthosis is: 4
The number of unique counts in feature erythema is: 4
The number of unique counts in feature family_history is: 2
The number of unique counts in feature hyperkeratosis is: 4
The number of unique counts in feature Age is: 60
The number of unique counts in feature inflammatory_monoluclear_inflitrate is: 4
The number of unique counts in feature eosinophils_in_the_infiltrate is: 3
The number of unique counts in feature class is: 6

# ===== Check values counts =====
final_df['Age'].value_counts()

# ===== Imputation =====
final_df['Age'] = final_df['Age'].replace(0, np.nan)

final_df['Age'].fillna(final_df['Age'].median(), inplace=True)

# ===== Checking the distribution and Q-Q plot of each continous variable from our final dataframe =====

# ===== Define continuous features to analyze =====
selected_features = ['Age']

# ===== Check skewness =====
print("Skewness Before Transformation:")
for col in selected_features:
    skew_val = round(final_df[col].skew(), 2)
    print(f"  {col}: {skew_val}")

# ===== Set theme =====
sns.set_style("darkgrid")

# ===== Plot Distribution + Q-Q side by side for each feature =====
for col in selected_features:
    fig, axes = plt.subplots(1, 2, figsize=(14, 4))

    # ===== Distribution plot (left) =====
    sns.histplot(final_df[col], kde=True, color='red', ax=axes[0])
    axes[0].set_title(f'Distribution of {col}')

    # ===== Q-Q plot (right) =====
    stats.probplot(final_df[col], dist="norm", plot=axes[1])
    axes[1].set_title(f'Q-Q Plot of {col}')

    # ===== Overall title for this feature only =====
    fig.suptitle(f"Analysis of {col}", fontsize=16, fontweight="bold", color="black", y=1.02)

    plt.tight_layout()
    plt.show()

Skewness Before Transformation:
  Age: 0.1

# ===== Applying StandardScaler for Feature Normalization =====

final_scale_df = final_df.copy()

scaler = StandardScaler()
final_scale_df['Age'] = scaler.fit_transform(final_scale_df[['Age']])

# ===== Split your data to train and test. Choose Splitting ratio wisely =====
x= final_scale_df.drop(columns='class',axis=1)
y= final_scale_df[['class']]

# ===== Spliting data =====
x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.2, random_state=0, stratify=y)

# ===== Checking the distribution of classes in training and testing sets =====
# Flatten y
y_train_flat = y_train.squeeze()
y_test_flat = y_test.squeeze()

# ===== Dataset Split Summary =====
split_summary = pd.DataFrame({
    "Dataset": ["x_train", "x_test", "y_train", "y_test"],
    "Shape": [x_train.shape, x_test.shape, y_train.shape, y_test.shape]
})

print("Dataset Split Summary\n")
print(split_summary.to_string(index=False))
print("-" * 53)

# ===== Target Variable Distribution (Counts & Percentages) =====
train_counts = pd.Series(y_train_flat).value_counts().rename("Train Count")
test_counts = pd.Series(y_test_flat).value_counts().rename("Test Count")

train_perc = (pd.Series(y_train_flat).value_counts(normalize=True)*100).round(2).rename("Train %")
test_perc = (pd.Series(y_test_flat).value_counts(normalize=True)*100).round(2).rename("Test %")

dist_summary = pd.concat([train_counts, test_counts, train_perc, test_perc], axis=1)
dist_summary.index.name = "Y"

print("\nTarget Variable Distribution (Counts & Percentages)\n")
print(dist_summary.to_string())

Dataset Split Summary

Dataset     Shape
x_train (292, 18)
 x_test  (74, 18)
y_train  (292, 1)
 y_test   (74, 1)
-----------------------------------------------------

Target Variable Distribution (Counts & Percentages)

   Train Count  Test Count  Train %  Test %
Y                                          
1           89          23    30.48   31.08
3           57          15    19.52   20.27
2           49          12    16.78   16.22
5           42          10    14.38   13.51
4           39          10    13.36   13.51
6           16           4     5.48    5.41

# ===== Fitting Logistic Regression Model =====
model_lgr = LogisticRegression(
    max_iter=500,
    class_weight='balanced',
    random_state=3
)

# ===== Train the model =====
start_time = time.time()
model_lgr.fit(x_train, y_train)
train_time = round(time.time() - start_time, 3)

# ===== Predictions =====
y_pred = model_lgr.predict(x_test)
y_pred_train = model_lgr.predict(x_train)

# ===== Analysing the model and evaluation Metric Score =====
# ===== Metrics =====
train_accuracy = accuracy_score(y_train, y_pred_train)
test_accuracy = accuracy_score(y_test, y_pred)
overfit_value = train_accuracy - test_accuracy
precision = precision_score(y_test, y_pred, average='macro', zero_division=0)
recall = recall_score(y_test, y_pred, average='macro', zero_division=0)
f1 = f1_score(y_test, y_pred, average='macro', zero_division=0)

# ===== Create DataFrame =====
results = {
    "Train Accuracy": round(train_accuracy, 4),
    "Test Accuracy": round(test_accuracy, 4),
    "Overfit Value": round(overfit_value, 4),
    "Precision (Macro)": round(precision, 4),
    "Recall (Macro)": round(recall, 4),
    "F1-Score (Macro)": round(f1, 4),
    "Training Time (sec)": train_time
}

results_df = pd.DataFrame(list(results.items()), columns=["Metric", "Value"])
# ===== Display =====
print("Logistic Regression Model Performance:")
print("\nConfusion Matrix:\n")
print(confusion_matrix(y_test, y_pred))
print("\nPerformance Metrics:\n")
print(results_df)

Logistic Regression Model Performance:

Confusion Matrix:

[[22  1  0  0  0  0]
 [ 0 10  1  0  0  1]
 [ 0  0 14  1  0  0]
 [ 0  1  2  7  0  0]
 [ 0  0  0  0 10  0]
 [ 0  1  0  0  0  3]]

Performance Metrics:

                Metric   Value
0       Train Accuracy  0.9555
1        Test Accuracy  0.8919
2        Overfit Value  0.0636
3    Precision (Macro)  0.8696
4       Recall (Macro)  0.8622
5     F1-Score (Macro)  0.8634
6  Training Time (sec)  0.0350

# ===== Fitting Random Forest Classifier Model =====
model_rf = RandomForestClassifier(
    n_estimators=100,           # number of trees
    max_depth=3,                # let it grow fully unless limited
    class_weight='balanced',    # handles imbalance
    random_state=1,
    n_jobs=-1                   # use all cores
)

# ===== Train the model =====
start_time = time.time()
model_rf.fit(x_train, y_train)
train_time = round(time.time() - start_time, 3)

# ===== Predictions =====
y_pred = model_rf.predict(x_test)
y_pred_train = model_rf.predict(x_train)

# ===== Analysing the model and evaluation Metric Score =====
# ===== Metrics =====
train_accuracy = accuracy_score(y_train, y_pred_train)
test_accuracy = accuracy_score(y_test, y_pred)
overfit_value = train_accuracy - test_accuracy
precision = precision_score(y_test, y_pred, average='macro', zero_division=0)
recall = recall_score(y_test, y_pred, average='macro', zero_division=0)
f1 = f1_score(y_test, y_pred, average='macro', zero_division=0)

# ===== Create DataFrame =====
results = {
    "Train Accuracy": round(train_accuracy, 4),
    "Test Accuracy": round(test_accuracy, 4),
    "Overfit Value": round(overfit_value, 4),
    "Precision (Macro)": round(precision, 4),
    "Recall (Macro)": round(recall, 4),
    "F1-Score (Macro)": round(f1, 4),
    "Training Time (sec)": train_time
}

results_df = pd.DataFrame(list(results.items()), columns=["Metric", "Value"])
# ===== Display =====
print("Random Forest Model Performance:")
print("\nConfusion Matrix:\n")
print(confusion_matrix(y_test, y_pred))
print("\nPerformance Metrics:\n")
print(results_df)

Random Forest Model Performance:

Confusion Matrix:

[[22  0  0  0  1  0]
 [ 0 11  0  0  0  1]
 [ 0  2  8  2  3  0]
 [ 0  0  1  8  1  0]
 [ 0  0  0  0 10  0]
 [ 0  0  0  0  1  3]]

Performance Metrics:

                Metric   Value
0       Train Accuracy  0.9144
1        Test Accuracy  0.8378
2        Overfit Value  0.0765
3    Precision (Macro)  0.8183
4       Recall (Macro)  0.8261
5     F1-Score (Macro)  0.8073
6  Training Time (sec)  0.2630

# ===== Fitting XGBoost Classifier Model =====
model_xgb = XGBClassifier(
    n_estimators=100,           # number of boosting rounds
    max_depth=3,                # depth of trees
    learning_rate=0.1,          # step size shrinkage
    subsample=0.8,              # row sampling
    colsample_bytree=0.8,       # feature sampling
    objective="multi:softmax",  # multi-class classification
    eval_metric="mlogloss",     # evaluation metric
    random_state=3,
    n_jobs=-1,
    use_label_encoder=False
)

# ===== Adjust labels to start from 0 =====
y_train_adj = y_train - 1
y_test_adj = y_test - 1

# ===== Train the model =====
start_time = time.time()
model_xgb.fit(x_train, y_train_adj)
train_time = round(time.time() - start_time, 3)

# ===== Predictions =====
y_pred = model_xgb.predict(x_test)
y_pred_train = model_xgb.predict(x_train)

# ===== Metrics =====
train_accuracy = accuracy_score(y_train_adj, y_pred_train)
test_accuracy = accuracy_score(y_test_adj, y_pred)
overfit_value = train_accuracy - test_accuracy
precision = precision_score(y_test_adj, y_pred, average='macro', zero_division=0)
recall = recall_score(y_test_adj, y_pred, average='macro', zero_division=0)
f1 = f1_score(y_test_adj, y_pred, average='macro', zero_division=0)

# ===== Create DataFrame =====
results = {
    "Train Accuracy": round(train_accuracy, 4),
    "Test Accuracy": round(test_accuracy, 4),
    "Overfit Value": round(overfit_value, 4),
    "Precision (Macro)": round(precision, 4),
    "Recall (Macro)": round(recall, 4),
    "F1-Score (Macro)": round(f1, 4),
    "Training Time (sec)": train_time
}

results_df = pd.DataFrame(list(results.items()), columns=["Metric", "Value"])

# ===== Display =====
print("XGBoost Model Performance:")
print("\nConfusion Matrix:\n")
print(confusion_matrix(y_test_adj, y_pred))
print("\nPerformance Metrics:\n")
print(results_df)

XGBoost Model Performance:

Confusion Matrix:

[[23  0  0  0  0  0]
 [ 0 11  0  0  1  0]
 [ 0  1 13  1  0  0]
 [ 0  2  1  7  0  0]
 [ 2  0  0  0  8  0]
 [ 1  1  0  0  0  2]]

Performance Metrics:

                Metric   Value
0       Train Accuracy  0.9966
1        Test Accuracy  0.8649
2        Overfit Value  0.1317
3    Precision (Macro)  0.8851
4       Recall (Macro)  0.7972
5     F1-Score (Macro)  0.8228
6  Training Time (sec)  0.1140

# ===== Fitting LightGBM Classifier Model =====
model_lgbm = LGBMClassifier(
    n_estimators=200,        # number of boosting iterations
    max_depth=4 ,            # no limit on depth
    learning_rate=0.5,       # shrinkage
    subsample=0.8,           # row sampling
    colsample_bytree=0.8,    # feature sampling
    objective="multiclass",  # multi-class classification
    random_state=3,
    n_jobs=-1
)

# ===== Adjust labels to start from 0 =====
y_train_adj = y_train - 1
y_test_adj = y_test - 1

# ===== Train the model =====
start_time = time.time()
model_lgbm.fit(x_train, y_train_adj)
train_time = round(time.time() - start_time, 3)

# ===== Predictions =====
y_pred = model_lgbm.predict(x_test)
y_pred_train = model_lgbm.predict(x_train)

# ===== Metrics =====
train_accuracy = accuracy_score(y_train_adj, y_pred_train)
test_accuracy = accuracy_score(y_test_adj, y_pred)
overfit_value = train_accuracy - test_accuracy
precision = precision_score(y_test_adj, y_pred, average='macro', zero_division=0)
recall = recall_score(y_test_adj, y_pred, average='macro', zero_division=0)
f1 = f1_score(y_test_adj, y_pred, average='macro', zero_division=0)

# ===== Create DataFrame =====
results = {
    "Train Accuracy": round(train_accuracy, 4),
    "Test Accuracy": round(test_accuracy, 4),
    "Overfit Value": round(overfit_value, 4),
    "Precision (Macro)": round(precision, 4),
    "Recall (Macro)": round(recall, 4),
    "F1-Score (Macro)": round(f1, 4),
    "Training Time (sec)": train_time
}

results_df = pd.DataFrame(list(results.items()), columns=["Metric", "Value"])

# ===== Display =====
print("LightGBM Model Performance:")
print("\nConfusion Matrix:\n")
print(confusion_matrix(y_test_adj, y_pred))
print("\nPerformance Metrics:\n")
print(results_df)

[LightGBM] [Info] Auto-choosing row-wise multi-threading, the overhead of testing was 0.000406 seconds.
You can set `force_row_wise=true` to remove the overhead.
And if memory is not enough, you can set `force_col_wise=true`.
[LightGBM] [Info] Total Bins 108
[LightGBM] [Info] Number of data points in the train set: 292, number of used features: 18
[LightGBM] [Info] Start training from score -1.188117
[LightGBM] [Info] Start training from score -1.784934
[LightGBM] [Info] Start training from score -1.633703
[LightGBM] [Info] Start training from score -2.013192
[LightGBM] [Info] Start training from score -1.939084
[LightGBM] [Info] Start training from score -2.904165
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
LightGBM Model Performance:

Confusion Matrix:

[[23  0  0  0  0  0]
 [ 0 12  0  0  0  0]
 [ 0  1 12  2  0  0]
 [ 0  2  1  7  0  0]
 [ 1  0  0  0  9  0]
 [ 0  0  0  0  2  2]]

Performance Metrics:

                Metric   Value
0       Train Accuracy  1.0000
1        Test Accuracy  0.8784
2        Overfit Value  0.1216
3    Precision (Macro)  0.8796
4       Recall (Macro)  0.8167
5     F1-Score (Macro)  0.8309
6  Training Time (sec)  0.1510

# ===== Hyperparameter Tuning - RandomForest Classifier =====
rf_hpt_pipeline = Pipeline([
    ('rf', RandomForestClassifier(class_weight='balanced', random_state=5, n_jobs=-1))
])

# ===== Hyperparameter grid =====
param_grid = {
    'rf__n_estimators': [100, 200, 300],
    'rf__max_depth': [3, 4, 5, 6],
    'rf__min_samples_split': [5, 10, 15, 20],
    'rf__min_samples_leaf': [3, 4, 5, 6],
    'rf__max_features': ['sqrt', 'log2']
}

# ===== GridSearchCV =====
grid_search = GridSearchCV(
    estimator=rf_hpt_pipeline,
    param_grid=param_grid,
    scoring='f1_macro',
    cv=5,
    n_jobs=-1,
    verbose=1
)

# ===== Train the model =====
start_time = time.time()
grid_search.fit(x_train, y_train)
train_time = round(time.time() - start_time, 3)

# ===== Best estimator =====
best_model = grid_search.best_estimator_

# ===== Predictions =====
y_pred = best_model.predict(x_test)
y_pred_train = best_model.predict(x_train)

# ===== Metrics =====
train_accuracy = accuracy_score(y_train, y_pred_train)
test_accuracy = accuracy_score(y_test, y_pred)
overfit_value = train_accuracy - test_accuracy
precision = precision_score(y_test, y_pred, average='macro', zero_division=0)
recall = recall_score(y_test, y_pred, average='macro', zero_division=0)
f1 = f1_score(y_test, y_pred, average='macro', zero_division=0)

# ===== Create DataFrame =====
results = {
    "Train Accuracy": round(train_accuracy, 4),
    "Test Accuracy": round(test_accuracy, 4),
    "Overfit Value": round(overfit_value, 4),
    "Precision (Macro)": round(precision, 4),
    "Recall (Macro)": round(recall, 4),
    "F1-Score (Macro)": round(f1, 4),
    "Training Time (sec)": train_time,
    "Best Params": grid_search.best_params_
}

results_df = pd.DataFrame(list(results.items()), columns=["Metric", "Value"])

# ===== Display Results =====
print("Random Forest Model Performance with Hyperparameter Tuning:")
print("\nConfusion Matrix:\n")
print(confusion_matrix(y_test, y_pred))
print("\nPerformance Metrics:\n")
print(results_df)

Fitting 5 folds for each of 384 candidates, totalling 1920 fits
Random Forest Model Performance with Hyperparameter Tuning:

Confusion Matrix:

[[22  1  0  0  0  0]
 [ 0 10  0  0  1  1]
 [ 0  2 11  2  0  0]
 [ 0  3  1  6  0  0]
 [ 0  0  0  0 10  0]
 [ 0  1  0  0  0  3]]

Performance Metrics:

                Metric                                              Value
0       Train Accuracy                                             0.9555
1        Test Accuracy                                             0.8378
2        Overfit Value                                             0.1176
3    Precision (Macro)                                              0.819
4       Recall (Macro)                                             0.8122
5     F1-Score (Macro)                                             0.8085
6  Training Time (sec)                                            834.682
7          Best Params  {'rf__max_depth': 6, 'rf__max_features': 'sqrt...

# ===== Convert features to float32 =====
x_train_adj = x_train.astype(np.float32)
x_test_adj = x_test.astype(np.float32)

# ===== Adjust labels to start from 0 =====
y_train_adj = y_train - 1
y_test_adj = y_test - 1

# ===== XGBoost pipeline =====
xgb_hpt_pipeline = Pipeline([
    ('xgb', XGBClassifier(
        objective="multi:softmax",
        eval_metric="mlogloss",
        use_label_encoder=False,
        random_state=5,
        n_jobs=-1
    ))
])

# ===== Hyperparameter grid =====
param_grid = {
    'xgb__n_estimators': [100, 200],
    'xgb__max_depth': [3, 4, 5],
    'xgb__learning_rate': [0.05, 0.1],
    'xgb__subsample': [0.8, 1.0],
    'xgb__colsample_bytree': [0.8, 1.0]
}


# ===== GridSearchCV =====
grid_search = GridSearchCV(
    estimator=xgb_hpt_pipeline,
    param_grid=param_grid,
    scoring='f1_macro',
    cv=5,
    n_jobs=-1,
    verbose=1,
    error_score='raise'
)

# ===== Train the model =====
start_time = time.time()
grid_search.fit(x_train_adj, y_train_adj)
train_time = round(time.time() - start_time, 3)

# ===== Best estimator =====
best_model = grid_search.best_estimator_

# ===== Predictions =====
y_pred = best_model.predict(x_test_adj)
y_pred_train = best_model.predict(x_train_adj)

# ===== Metrics =====
train_accuracy = accuracy_score(y_train_adj, y_pred_train)
test_accuracy = accuracy_score(y_test_adj, y_pred)
overfit_value = train_accuracy - test_accuracy
precision = precision_score(y_test_adj, y_pred, average='macro', zero_division=0)
recall = recall_score(y_test_adj, y_pred, average='macro', zero_division=0)
f1 = f1_score(y_test_adj, y_pred, average='macro', zero_division=0)

# ===== Create DataFrame =====
results = {
    "Train Accuracy": round(train_accuracy, 4),
    "Test Accuracy": round(test_accuracy, 4),
    "Overfit Value": round(overfit_value, 4),
    "Precision (Macro)": round(precision, 4),
    "Recall (Macro)": round(recall, 4),
    "F1-Score (Macro)": round(f1, 4),
    "Training Time (sec)": train_time,
    "Best Params": grid_search.best_params_
}

results_df = pd.DataFrame(list(results.items()), columns=["Metric", "Value"])

# ===== Display Results =====
print("XGBoost Model Performance with Hyperparameter Tuning:")
print("\nConfusion Matrix:\n")
print(confusion_matrix(y_test_adj, y_pred))
print("\nPerformance Metrics:\n")
print(results_df)

Fitting 5 folds for each of 48 candidates, totalling 240 fits
XGBoost Model Performance with Hyperparameter Tuning:

Confusion Matrix:

[[23  0  0  0  0  0]
 [ 0 10  0  0  1  1]
 [ 0  1 13  1  0  0]
 [ 0  2  1  7  0  0]
 [ 2  0  0  0  8  0]
 [ 1  0  0  0  1  2]]

Performance Metrics:

                Metric                                              Value
0       Train Accuracy                                             0.9863
1        Test Accuracy                                             0.8514
2        Overfit Value                                              0.135
3    Precision (Macro)                                             0.8207
4       Recall (Macro)                                             0.7833
5     F1-Score (Macro)                                             0.7974
6  Training Time (sec)                                             34.693
7          Best Params  {'xgb__colsample_bytree': 0.8, 'xgb__learning_...

# ===== Convert features to float32 =====
x_train_adj = x_train.astype(np.float32)
x_test_adj = x_test.astype(np.float32)

# ===== Adjust labels to start from 0 =====
y_train_adj = y_train - 1
y_test_adj = y_test - 1

# ===== LightGBM pipeline =====
lgbm_hpt_pipeline = Pipeline([
    ('lgbm', LGBMClassifier(
        objective="multiclass",
        random_state=5,
        n_jobs=-1
    ))
])

# ===== Hyperparameter grid =====
param_grid = {
    'lgbm__n_estimators': [100, 200],          # more boosting rounds
    'lgbm__max_depth': [2, 3, 4],              # shallower trees
    'lgbm__learning_rate': [0.01, 0.05],       # slower learning
    'lgbm__subsample': [0.6, 0.8],             # row sampling
    'lgbm__colsample_bytree': [0.6, 0.8],      # feature sampling
    'lgbm__min_child_samples': [10, 20],       # prevent overfitting
}

# ===== RandomizedSearchCV =====
grid_search = RandomizedSearchCV(
    estimator=lgbm_hpt_pipeline,
    param_distributions=param_grid,
    n_iter=15,
    scoring='f1_macro',
    cv=2,
    n_jobs=-1,
    verbose=1,
    random_state=42,
    error_score='raise'
)

# ===== Train the model =====
start_time = time.time()
grid_search.fit(x_train_adj, y_train_adj)
train_time = round(time.time() - start_time, 3)

# ===== Best estimator =====
best_model = grid_search.best_estimator_

# ===== Predictions =====
y_pred = best_model.predict(x_test_adj)
y_pred_train = best_model.predict(x_train_adj)

# ===== Metrics =====
train_accuracy = accuracy_score(y_train_adj, y_pred_train)
test_accuracy = accuracy_score(y_test_adj, y_pred)
overfit_value = train_accuracy - test_accuracy
precision = precision_score(y_test_adj, y_pred, average='macro', zero_division=0)
recall = recall_score(y_test_adj, y_pred, average='macro', zero_division=0)
f1 = f1_score(y_test_adj, y_pred, average='macro', zero_division=0)

# ===== Create DataFrame =====
results = {
    "Train Accuracy": round(train_accuracy, 4),
    "Test Accuracy": round(test_accuracy, 4),
    "Overfit Value": round(overfit_value, 4),
    "Precision (Macro)": round(precision, 4),
    "Recall (Macro)": round(recall, 4),
    "F1-Score (Macro)": round(f1, 4),
    "Training Time (sec)": train_time,
    "Best Params": grid_search.best_params_
}

results_df = pd.DataFrame(list(results.items()), columns=["Metric", "Value"])

# ===== Display Results =====
print("LightGBM Model Performance with Hyperparameter Tuning:")
print("\nConfusion Matrix:\n")
print(confusion_matrix(y_test_adj, y_pred))
print("\nPerformance Metrics:\n")
print(results_df)

Fitting 2 folds for each of 15 candidates, totalling 30 fits
[LightGBM] [Info] Auto-choosing row-wise multi-threading, the overhead of testing was 0.000130 seconds.
You can set `force_row_wise=true` to remove the overhead.
And if memory is not enough, you can set `force_col_wise=true`.
[LightGBM] [Info] Total Bins 108
[LightGBM] [Info] Number of data points in the train set: 292, number of used features: 18
[LightGBM] [Info] Start training from score -1.188117
[LightGBM] [Info] Start training from score -1.784934
[LightGBM] [Info] Start training from score -1.633703
[LightGBM] [Info] Start training from score -2.013192
[LightGBM] [Info] Start training from score -1.939084
[LightGBM] [Info] Start training from score -2.904165
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
LightGBM Model Performance with Hyperparameter Tuning:

Confusion Matrix:

[[23  0  0  0  0  0]
 [ 0  9  0  0  2  1]
 [ 0  1 12  1  1  0]
 [ 0  2  1  7  0  0]
 [ 2  0  0  0  8  0]
 [ 1  0  0  0  1  2]]

Performance Metrics:

                Metric                                              Value
0       Train Accuracy                                             0.9829
1        Test Accuracy                                             0.8243
2        Overfit Value                                             0.1586
3    Precision (Macro)                                             0.7943
4       Recall (Macro)                                             0.7583
5     F1-Score (Macro)                                             0.7704
6  Training Time (sec)                                             62.881
7          Best Params  {'lgbm__subsample': 0.6, 'lgbm__n_estimators':...

# ===== Store results =====
results = {
    "Logistic Regression": {
        'Training Accuracy': 0.9555,
        'Test Accuracy': 0.8919,
        'Overfit (Train - Test Acc)': 0.0636,
        'Precision': 0.8696,
        'Recall': 0.8622,
        'F1-Score': 0.8634,
        'Training Time (sec)': 0.3350
    },
    "Random Forest": {
        'Training Accuracy': 0.9144,
        'Test Accuracy': 0.8378,
        'Overfit (Train - Test Acc)': 0.0765,
        'Precision': 0.8183,
        'Recall': 0.8261,
        'F1-Score': 0.8073,
        'Training Time (sec)': 0.5600
    },
    "XGBoost": {
        'Training Accuracy': 0.9966,
        'Test Accuracy': 0.8649,
        'Overfit (Train - Test Acc)': 0.1317,
        'Precision': 0.8851,
        'Recall': 0.7972,
        'F1-Score': 0.8228,
        'Training Time (sec)': 0.9150
    },
    "LightGBM": {
        'Training Accuracy': 1.0000,
        'Test Accuracy': 0.8784,
        'Overfit (Train - Test Acc)': 0.1216,
        'Precision': 0.8796,
        'Recall': 0.8167,
        'F1-Score': 0.8309,
        'Training Time (sec)': 0.7670
    }
}

# ===== Convert to DataFrame =====
df_results = pd.DataFrame(results).T

print("\n=== Model Comparison Table ===")
df_results

=== Model Comparison Table ===

# ===== Comparing Model Performance Scores =====
def add_labels(ax, decimals=3, threshold=0.05):
    """Add value labels on bars dynamically adjusting position and color."""
    y_lim = ax.get_ylim()[1]
    for p in ax.patches:
        value = p.get_height()
        bar_height_ratio = abs(value) / y_lim

        if bar_height_ratio > threshold:
            y = value - (y_lim * 0.02)
            va = 'top'
            color = "black"
        else:
            y = value + (y_lim * 0.01)
            va = 'bottom'
            color = "black"

        ax.annotate(f"{value:.{decimals}f}",
                    (p.get_x() + p.get_width() / 2., y),
                    ha='center', va=va, fontsize=9,
                    color=color, fontweight="bold", rotation=90)

# ===== 1. Metrics on 0–1 scale =====
metrics1 = ["Training Accuracy", "Test Accuracy", "Precision", "Recall", "F1-Score"]
plot_df1 = df_results[metrics1]

ax1 = plot_df1.plot(kind='bar', figsize=(20, 4), width=0.8, colormap="Blues")
plt.title("Model Performance (Accuracy, Precision, Recall, F1)", fontsize=16, fontweight='bold')
plt.ylabel("Score", fontsize=12)
plt.xticks(rotation=0, fontsize=11)
plt.legend(title="Metrics", bbox_to_anchor=(1.05, 1), loc='upper left')
plt.grid(axis='y', linestyle='--', alpha=0.7)
add_labels(ax1, decimals=3)
plt.tight_layout()
plt.show()

# ===== 2. Training time only =====
metrics2 = ["Training Time (sec)"]
plot_df2 = df_results[metrics2]

ax2 = plot_df2.plot(kind='bar', figsize=(15, 4), width=0.6, colormap="plasma")
plt.title("Model Training Time", fontsize=16, fontweight='bold')
plt.ylabel("Seconds", fontsize=12)
plt.xticks(rotation=0, fontsize=11)
plt.legend(title="Metrics", bbox_to_anchor=(1.05, 1), loc='upper left')
plt.grid(axis='y', linestyle='--', alpha=0.7)
add_labels(ax2, decimals=3)
plt.tight_layout()
plt.show()

# ===== Comparing Model Accuracy Scores =====
def add_value_labels(ax, decimals=3, threshold=0.05):
    x_lim = ax.get_xlim()[1]
    for p in ax.patches:
        value = p.get_width()
        bar_width_ratio = abs(value) / x_lim
        if bar_width_ratio > threshold:
            x = value - (x_lim * 0.02)
            ha = 'right'
            color = "white"
        else:
            x = value + (x_lim * 0.01)
            ha = 'left'
            color = "black"
        txt = ax.annotate(f"{value:.{decimals}f}",
                          (x, p.get_y() + p.get_height() / 2.),
                          va='center', ha=ha, fontsize=10,
                          color=color, fontweight="bold")
        txt.set_path_effects([
            path_effects.Stroke(linewidth=2, foreground='black'),
            path_effects.Normal()
        ])

# ===== Accuracy =====
metrics3 = ["Test Accuracy"]
plot_df3 = df_results[metrics3]
ax = plot_df3.plot(kind='barh', figsize=(9, 4), width=0.6,
                  color="red", edgecolor="black")
plt.title("Model Accuracy", fontsize=16, fontweight='bold', color="navy")
plt.xlabel("Accuracy Score", fontsize=12)
plt.yticks(fontsize=11, fontweight="bold")
plt.grid(axis='x', linestyle='--', alpha=0.7)
add_value_labels(ax, decimals=3)
plt.tight_layout()
plt.show()

# ===== Store results =====
results_2 = {
    "Random Forest Classifier": {
        'Training Accuracy': 0.9555,
        'Test Accuracy': 0.8378,
        'Overfit (Train - Test Acc)': 0.1176,
        'Precision': 0.8190,
        'Recall': 0.8122,
        'F1-Score': 0.8085,
        'Training Time (sec)': 680.123
    },

    "XGBoost Classifier": {
        'Training Accuracy': 0.9863,
        'Test Accuracy': 0.8514,
        'Overfit (Train - Test Acc)': 0.1350,
        'Precision': 0.8207,
        'Recall': 0.7833,
        'F1-Score': 0.7974,
        'Training Time (sec)': 37.716
    },

    "LightGBM Classifier": {
        'Training Accuracy': 1.0000,
        'Test Accuracy': 0.8514,
        'Overfit (Train - Test Acc)': 0.1486,
        'Precision': 0.8108,
        'Recall': 0.8139,
        'F1-Score': 0.8088,
        'Training Time (sec)': 84.086
    }
}

# ===== Convert to DataFrame =====
df_results_2 = pd.DataFrame(results_2).T
print("\n=== Hyperparameter-Tuning Comparison Table ===")
df_results_2

=== Hyperparameter-Tuning Comparison Table ===

# ===== Comparing Hyperparameter-Tuning Performance Scores =====
def add_labels(ax, decimals=3, threshold=0.05):

    y_lim = ax.get_ylim()[1]
    for p in ax.patches:
        value = p.get_height()
        bar_height_ratio = abs(value) / y_lim

        if bar_height_ratio > threshold:
            y = value - (y_lim * 0.02)
            va = 'top'
            color = "white"
        else:
            y = value + (y_lim * 0.01)
            va = 'bottom'
            color = "black"

        ax.annotate(f"{value:.{decimals}f}",
                    (p.get_x() + p.get_width() / 2., y),
                    ha='center', va=va, fontsize=9,
                    color='black', fontweight="bold", rotation=90)

# ===== 1. Metrics on 0–1 scale =====
metrics4 = ["Training Accuracy", "Test Accuracy", "Precision", "Recall", "F1-Score"]
plot_df4 = df_results_2[metrics4]

ax1 = plot_df4.plot(kind='bar', figsize=(20, 6), width=0.8, colormap="Reds")
plt.title("Model Performance (Accuracy, Precision, Recall, F1)", fontsize=16, fontweight='bold')
plt.ylabel("Score", fontsize=12)
plt.xticks(rotation=0, fontsize=11)
plt.legend(title="Metrics", bbox_to_anchor=(1.05, 1), loc='upper left')
plt.grid(axis='y', linestyle='--', alpha=0.7)
add_labels(ax1, decimals=3)
plt.tight_layout()
plt.show()

# ===== Comparing Hyperparameter-Tuning Accuracy Scores =====
def add_value_labels(ax, decimals=3, threshold=0.05):
    x_lim = ax.get_xlim()[1]
    for p in ax.patches:
        value = p.get_width()
        bar_width_ratio = abs(value) / x_lim
        if bar_width_ratio > threshold:
            x = value - (x_lim * 0.02)
            ha = 'right'
            color = "white"
        else:
            x = value + (x_lim * 0.01)
            ha = 'left'
            color = "black"
        txt = ax.annotate(f"{value:.{decimals}f}",
                          (x, p.get_y() + p.get_height() / 2.),
                          va='center', ha=ha, fontsize=10,
                          color=color, fontweight="bold")
        txt.set_path_effects([
            path_effects.Stroke(linewidth=2, foreground='black'),
            path_effects.Normal()
        ])

# ===== Accuracy =====
metrics6 = ["Test Accuracy"]
plot_df6 = df_results_2[metrics6]
ax = plot_df6.plot(kind='barh', figsize=(9, 5), width=0.6,
                  color="navy", edgecolor="black")
plt.title("Hyperparameter-Tuning Accuracy", fontsize=16, fontweight='bold', color="#641E16")
plt.xlabel("Accuracy Score", fontsize=12)
plt.yticks(fontsize=11, fontweight="bold")
plt.grid(axis='x', linestyle='--', alpha=0.7)
add_value_labels(ax, decimals=3)
plt.tight_layout()
plt.show()

# ===== Define CV strategy =====
cv = 5
skf = StratifiedKFold(n_splits=cv, shuffle=True, random_state=42)

# ===== Encode y_train to start from 0 =====
le = LabelEncoder()
y_train_enc = le.fit_transform(y_train)

# ===== Dictionary of models =====
models = {
    "Logistic Regression": model_lgr,
    "Random Forest": model_rf,
    "XGBoost": model_xgb,
    "LightGBM": model_lgbm
}

# ===== Store results =====
results = {}
for name, model in models.items():
    scores = cross_val_score(model, x_train, y_train_enc, cv=skf,
                             scoring='accuracy', n_jobs=-1)
    results[name] = scores.mean()
    print(f"{name} - CV Accuracy Scores: {scores}")
    print(f"{name} - Mean CV Accuracy: {scores.mean():.4f}\n")

# ===== Convert results to DataFrame =====
df_cv_results = pd.DataFrame(list(results.items()), columns=["Model", "Mean CV Accuracy"])
print("\n=== Cross-Validation Comparison Table ===")
print(df_cv_results)

Logistic Regression - CV Accuracy Scores: [0.93220339 0.88135593 0.86206897 0.84482759 0.87931034]
Logistic Regression - Mean CV Accuracy: 0.8800

Random Forest - CV Accuracy Scores: [0.84745763 0.86440678 0.89655172 0.82758621 0.86206897]
Random Forest - Mean CV Accuracy: 0.8596

XGBoost - CV Accuracy Scores: [0.86440678 0.89830508 0.9137931  0.82758621 0.87931034]
XGBoost - Mean CV Accuracy: 0.8767

LightGBM - CV Accuracy Scores: [0.84745763 0.91525424 0.87931034 0.87931034 0.89655172]
LightGBM - Mean CV Accuracy: 0.8836


=== Cross-Validation Comparison Table ===
                 Model  Mean CV Accuracy
0  Logistic Regression          0.879953
1        Random Forest          0.859614
2              XGBoost          0.876680
3             LightGBM          0.883577

# ===== Sort values for better visualization =====
df_cv_results = df_cv_results.sort_values(by="Mean CV Accuracy", ascending=True)

# ===== Plot =====
plt.figure(figsize=(9,4))
sns.barplot(
    data=df_cv_results,
    x="Mean CV Accuracy",
    y="Model",
    color="green",
    edgecolor="black"
)

# ===== Add accuracy values on bars =====
for i, v in enumerate(df_cv_results["Mean CV Accuracy"]):
    plt.text(v + 0.002, i, f"{v:.3f}", va="center", fontweight="bold")

plt.title("Model Comparison - Mean CV Accuracy", fontsize=16, fontweight="bold", color='red')
plt.xlabel("Mean CV Accuracy")
plt.ylabel("Model")
plt.xlim(0, 1)
plt.show()

# ===== Comparison For ML Model Accuracy vs Hyperparameter-Tuning Accuracy vs CV Accuracy =====
# ===== Train Accuracy =====
train_acc = {
    "Logistic Regression": 0.9555,
    "Random Forest": 0.9144,
    "XGBoost": 0.9966,
    "LightGBM": 1.0000
}

train_tuned_acc = {
    "Random Forest": 0.9555,
    "XGBoost": 0.9863,
    "LightGBM": 1.0000
}

# ===== Test Accuracy =====
test_acc = {
    "Logistic Regression": 0.8919,
    "Random Forest": 0.8378,
    "XGBoost": 0.8649,
    "LightGBM": 0.8784
}

tuned_acc = {
    "Random Forest": 0.8378,
    "XGBoost": 0.8514,
    "LightGBM": 0.8514
}

# ===== Cross-Validation Mean Accuracy =====
cv_acc = {
    "Logistic Regression": 0.879953,
    "Random Forest": 0.859614,
    "XGBoost": 0.876680,
    "LightGBM": 0.883577
}

# ===== Combine into DataFrame =====
df_compare = pd.DataFrame({
    "Model": list(set(list(train_acc.keys()) +
                      list(train_tuned_acc.keys()) +
                      list(test_acc.keys()) +
                      list(tuned_acc.keys()) +
                      list(cv_acc.keys())))
})

df_compare["Train Accuracy (Before Tuning)"] = df_compare["Model"].map(train_acc)
df_compare["Train Accuracy (After Tuning)"] = df_compare["Model"].map(train_tuned_acc)
df_compare["Test Accuracy (Before Tuning)"] = df_compare["Model"].map(test_acc)
df_compare["Test Accuracy (After Tuning)"] = df_compare["Model"].map(tuned_acc)
df_compare["CV Accuracy"] = df_compare["Model"].map(cv_acc)

# ===== Melt for grouped bar chart =====
df_melted = df_compare.melt(
    id_vars="Model",
    var_name="Metric",
    value_name="Accuracy"
)

# ===== Drop NaN rows =====
df_melted = df_melted.dropna(subset=["Accuracy"])

# ===== Custom colors mapping =====
custom_palette = {
    "Train Accuracy (Before Tuning)": "#FFD700",
    "Train Accuracy (After Tuning)": "#800000",
    "Test Accuracy (Before Tuning)": "navy",
    "Test Accuracy (After Tuning)": "red",
    "CV Accuracy": "purple"
}

# ===== Plot =====
plt.figure(figsize=(20,8))
ax = sns.barplot(
    data=df_melted,
    x="Model", y="Accuracy", hue="Metric",
    palette=custom_palette
)

# ===== Annotate bars =====
for p in ax.patches:
    height = p.get_height()
    if height > 0:
        ax.annotate(f"{height:.3f}",
                    (p.get_x() + p.get_width() / 2., height),
                    ha='center', va='bottom', fontsize=9, color='black', xytext=(0,2), textcoords='offset points')

plt.title("ML Model Performance: Train vs Test Accuracy (Before & After Tuning) vs CV Accuracy",
          fontsize=16, fontweight="bold", pad=23)
plt.ylabel("Accuracy")
plt.ylim(0,1)

# ===== Legend =====
plt.legend(title="Metric",
           bbox_to_anchor=(1.05, 1),
           loc='upper left')
plt.tight_layout()
plt.show()

# ===== Create Pipeline =====
final_model_lgr_pipeline = Pipeline([
    ('scaler', StandardScaler()),
    ('classifier', LogisticRegression(
        max_iter=500,
        class_weight='balanced',
        random_state=3
    ))
])

# ===== Fit the pipeline =====
final_model_lgr_pipeline.fit(x_train, y_train)

Pipeline(steps=[('scaler', StandardScaler()),
                ('classifier',
                 LogisticRegression(class_weight='balanced', max_iter=500,
                                    random_state=3))])

Pipeline(steps=[('scaler', StandardScaler()),
                ('classifier',
                 LogisticRegression(class_weight='balanced', max_iter=500,
                                    random_state=3))])

StandardScaler()

LogisticRegression(class_weight='balanced', max_iter=500, random_state=3)

# ===== Make predictions on test set =====
y_pred = final_model_lgr_pipeline.predict(x_test)

# ===== Classification Report =====
report = classification_report(y_test, y_pred)
print(report)

# ===== Confusion Matrix =====
cm = confusion_matrix(y_test, y_pred)
print("Confusion Matrix:\n", cm)

              precision    recall  f1-score   support

           1       1.00      1.00      1.00        23
           2       0.77      0.83      0.80        12
           3       0.82      0.93      0.88        15
           4       0.88      0.70      0.78        10
           5       1.00      0.90      0.95        10
           6       0.75      0.75      0.75         4

    accuracy                           0.89        74
   macro avg       0.87      0.85      0.86        74
weighted avg       0.90      0.89      0.89        74

Confusion Matrix:
 [[23  0  0  0  0  0]
 [ 0 10  1  0  0  1]
 [ 0  0 14  1  0  0]
 [ 0  1  2  7  0  0]
 [ 0  1  0  0  9  0]
 [ 0  1  0  0  0  3]]

# ===== Predict on training data =====
y_train_pred = final_model_lgr_pipeline.predict(x_train)
train_acc = accuracy_score(y_train, y_train_pred)
print("=== Training Accuracy ===")
print(f"Training Accuracy: {train_acc:.4f}\n")

# ===== Predict on test data =====
y_test_pred = final_model_lgr_pipeline.predict(x_test)
test_acc = accuracy_score(y_test, y_test_pred)
print("=== Testing Accuracy ===")
print(f"Testing Accuracy: {test_acc:.4f}\n")

=== Training Accuracy ===
Training Accuracy: 0.9623

=== Testing Accuracy ===
Testing Accuracy: 0.8919

# ===== Compute macro-average ROC curve and AUC =====
classes = np.unique(y_test)
y_test_bin = label_binarize(y_test, classes=classes)
n_classes = y_test_bin.shape[1]

# ===== Get predicted probabilities =====
y_score = final_model_lgr_pipeline.predict_proba(x_test)

# ===== Compute ROC curve and AUC for each class =====
fpr, tpr, roc_auc = {}, {}, {}
for i in range(n_classes):
    fpr[i], tpr[i], _ = roc_curve(y_test_bin[:, i], y_score[:, i])
    roc_auc[i] = auc(fpr[i], tpr[i])

# ===== Compute micro-average ROC curve and AUC =====
fpr["micro"], tpr["micro"], _ = roc_curve(y_test_bin.ravel(), y_score.ravel())
roc_auc["micro"] = auc(fpr["micro"], tpr["micro"])

# ===== Aggregate all false positive rates =====
all_fpr = np.unique(np.concatenate([fpr[i] for i in range(n_classes)]))
# ===== Interpolate all ROC curves at these points =====
mean_tpr = np.zeros_like(all_fpr)
for i in range(n_classes):
    mean_tpr += np.interp(all_fpr, fpr[i], tpr[i])
mean_tpr /= n_classes
fpr["macro"] = all_fpr
tpr["macro"] = mean_tpr
roc_auc["macro"] = auc(fpr["macro"], tpr["macro"])

# ===== Style settings =====
plt.figure(figsize=(20, 6))
sns.set_style("whitegrid")
colors = sns.color_palette("tab10", n_classes)

# ===== Create axis with gray background =====
ax = plt.gca()
ax.set_facecolor("#f0f0f0")
for i, color in zip(range(n_classes), colors):
    plt.plot(
        fpr[i], tpr[i],
        color=color, lw=2,
        label=f"Class {classes[i]} (AUC = {roc_auc[i]:.2f})"
    )

plt.plot(
    fpr["micro"], tpr["micro"],
    color="black", linestyle=":", linewidth=3,
    label=f"Micro-average (AUC = {roc_auc['micro']:.2f})"
)

plt.plot(
    fpr["macro"], tpr["macro"],
    color="navy", linestyle="--", linewidth=3,
    label=f"Macro-average (AUC = {roc_auc['macro']:.2f})"
)

plt.plot([0, 1], [0, 1], "k--", lw=2, alpha=0.6)

plt.xlim([0.0, 1.0])
plt.ylim([0.0, 1.05])
plt.xlabel("False Positive Rate", fontsize=14, fontweight="bold")
plt.ylabel("True Positive Rate", fontsize=14, fontweight="bold")
plt.title("Multiclass ROC–AUC Curve (Logistic Regression Pipeline)",
          fontsize=16, fontweight="bold", pad=20)
plt.legend(
    loc="center left",
    bbox_to_anchor=(1, 0.5),
    fontsize=12,
    frameon=True,
    shadow=True
)
plt.tight_layout()
plt.show()

# ===== Checking the percentage of feature importance =====
features = final_scale_df.drop(columns=[final_scale_df.columns[-1]]).columns
coefficients = final_model_lgr_pipeline.named_steps['classifier'].coef_
importance = np.mean(np.abs(coefficients), axis=0)
feature_imp = pd.DataFrame({'Variable': features, 'Importance': importance})
feature_imp['Importance (%)'] = (feature_imp['Importance'] / feature_imp['Importance'].sum() * 100).round(2)
feature_imp = feature_imp.sort_values(by='Importance (%)', ascending=False).reset_index(drop=True)
print(feature_imp[['Variable', 'Importance (%)']])

                               Variable  Importance (%)
0                    koebner_phenomenon           10.55
1                               itching            9.72
2                            spongiosis            8.72
3                                   Age            8.15
4                        PNL_infiltrate            7.88
5   disappearance_of_the_granular_layer            6.59
6                               scaling            6.49
7                            exocytosis            6.13
8                         parakeratosis            5.18
9                            acanthosis            4.68
10                    munro_microabcess            4.48
11        eosinophils_in_the_infiltrate            4.05
12                     definite_borders            3.87
13  inflammatory_monoluclear_inflitrate            3.53
14                   spongiform_pustule            3.17
15                       family_history            2.75
16                       hyperkeratosis            2.42
17                             erythema            1.64

# ===== Plotting the barplot to determine which feature is contributing the most =====
plt.figure(figsize=(20,7))
plt.gcf().set_facecolor('#f2f2f2')
sns.set_style("whitegrid", {"axes.facecolor": "#e6e6e6"})
colors = sns.color_palette("viridis", n_colors=len(feature_imp))

# ===== Use the correct column names =====
barplot = sns.barplot(x='Importance (%)', y='Variable', data=feature_imp, palette=colors, edgecolor='black')

# ===== Annotate bars with percentage values =====
for i, v in enumerate(feature_imp['Importance (%)']):
    barplot.text(v + 0.5, i, f"{v:.2f}%", va='center', fontsize=10, fontweight='bold')

plt.title('Feature Importances (Logistic Regression)', fontsize=20, fontweight='bold', color="#333333", pad=20)
plt.xlabel('Importance (%)', fontsize=14, fontweight='bold', color="#333333")
plt.ylabel('Features', fontsize=14, fontweight='bold', color="#333333")
plt.grid(axis='x', linestyle='--', alpha=0.6)
plt.tight_layout()
plt.show()

# ===== Importing pickle module =====
import pickle

# ===== Define model and path =====
model = final_model_lgr_pipeline

# ===== Save model using pickle =====
with open("SkinDisorder.pkl", "wb") as f:
    pickle.dump(model, f)

print("Model saved successfully as 'SkinDisorder.pkl'")

Model saved successfully as 'SkinDisorder.pkl'

# ===== Load the File and predict unseen data =====
# ===== Load the model in read-binary ('rb') mode =====
with open("SkinDisorder.pkl", "rb") as f:
    lgr_model = pickle.load(f)

# ===== Predict on unseen (test) data =====
predictions = lgr_model.predict(x_test)

# ===== Display predictions =====
print("Predictions on test data:")
print(predictions)

# ===== Evaluate =====
print("\nLightGBM Classification Report:\n")
print(classification_report(y_test, predictions))

Predictions on test data:
[1 6 1 2 1 1 2 1 2 5 2 1 1 4 1 1 1 5 5 1 1 1 3 2 3 5 4 1 6 3 4 1 5 2 5 1 2
 3 3 4 4 3 3 3 2 3 1 1 3 1 5 3 2 3 3 5 4 1 3 1 6 4 2 2 2 3 1 5 1 2 6 3 4 3]

LightGBM Classification Report:

              precision    recall  f1-score   support

           1       1.00      1.00      1.00        23
           2       0.77      0.83      0.80        12
           3       0.82      0.93      0.88        15
           4       0.88      0.70      0.78        10
           5       1.00      0.90      0.95        10
           6       0.75      0.75      0.75         4

    accuracy                           0.89        74
   macro avg       0.87      0.85      0.86        74
weighted avg       0.90      0.89      0.89        74

No.	Feature	Description
1	Erythema	Presence or absence of redness or inflammation of the skin.
2	Scaling	Whether there are visible scales or flakes on the skin's surface.
3	Definite Borders	Describes whether the skin lesion has well-defined or clear borders.
4	Itching	Indicates if the patient experiences itching (pruritus).
5	Koebner Phenomenon	Whether new lesions appear at sites of trauma or injury to the skin.
6	Polygonal Papules	Presence of polygonal-shaped raised skin lesions.
7	Follicular Papules	Raised lesions involving hair follicles.
8	Oral Mucosal Involvement	Whether the condition affects mucous membranes inside the mouth.
9	Knee and Elbow Involvement	Whether the condition is specifically located on knees and elbows.
10	Scalp Involvement	Whether the condition affects the scalp.
11	Family History (0 or 1)	Indicates presence of family history of similar conditions (genetic factors).
12	Age	Age of the patient at the time of data collection.

No.	Feature	Description
1	Melanin Incontinence	Presence of melanin pigment within the dermis due to melanocyte damage or melanin leakage.
2	Eosinophils in the Infiltrate	Indicates eosinophils (allergy/inflammation cells) in the inflammatory infiltrate.
3	PNL Infiltrate	Presence of polymorphonuclear leukocytes, linked to early-stage inflammation.
4	Fibrosis of the Papillary Dermis	Excess fibrous tissue in the papillary dermis, usually from chronic inflammation/scarring.
5	Exocytosis	Migration of inflammatory cells (e.g., lymphocytes) into the epidermis.
6	Acanthosis	Thickening of the epidermis (stratum spinosum), seen in psoriasis and acanthosis nigricans.
7	Hyperkeratosis	Excessive thickening of the outer skin layer (stratum corneum).
8	Parakeratosis	Retention of nuclei in the stratum corneum, common in psoriasis.
9	Clubbing of the Rete Ridges	Bulbous enlargement of rete ridges (epidermal projections into dermis).
10	Elongation of the Rete Ridges	Lengthening of epidermal projections into the dermis.
11	Thinning of Suprapapillary Epidermis	Reduced thickness above dermal papillae, seen in some inflammatory disorders.
12	Spongiform Pustule	Blister-like epidermal structure filled with neutrophils; hallmark of pustular psoriasis.
13	Munro Microabscess	Small clusters of neutrophils in the stratum corneum, typical of psoriasis.
14	Focal Hypergranulosis	Localized thickening of the granular epidermal layer.
15	Disappearance of Granular Layer	Loss or reduction of the epidermal granular layer.
16	Vacuolization of Basal Layer	Empty vacuoles in the basal epidermis, linked to autoimmune blistering disorders.
17	Spongiosis	Fluid buildup between epidermal cells (intercellular edema), common in eczema.
18	Saw-tooth Appearance of Rete Ridges	Jagged, irregular epidermal projections, common in lichen planus.
19	Follicular Horn Plug	Keratin-filled plug inside a hair follicle (e.g., acne).
20	Perifollicular Parakeratosis	Parakeratosis specifically around hair follicles.
21	Inflammatory Mononuclear Infiltrate	Mononuclear WBCs (e.g., lymphocytes) in dermal/epidermal tissue, sign of chronic inflammation.
22	Band-like Infiltrate	Dense, linear accumulation of inflammatory cells, often in lichen planus.

Class No.	Disease Name	Description
1	Psoriasis	A skin disease that causes itchy, scaly rashes, most commonly on knees, elbows, trunk, and scalp.
2	Seboreic Dermatitis	A common inflammatory condition mainly affecting the scalp, causing dandruff, scaly patches, and inflamed skin; also affects oily areas (face, nose, eyebrows, ears, chest).
3	Lichen Planus	An inflammatory condition affecting skin, hair, nails, and mucous membranes; appears as purple, itchy, flat bumps on skin and lacy white patches in mouth/genitals.
4	Pityriasis Rosea	A rash beginning with a “herald patch” (oval spot), followed by smaller spots in a pine-tree pattern; itchy, self-limiting, and most common between ages 10–35.
5	Chronic Dermatitis	A long-term condition causing swelling, irritation, itchy/dry skin, rashes, or blisters. Includes atopic dermatitis (eczema), contact dermatitis, and seborrheic dermatitis.
6	Pityriasis Rubra Pilaris	A rare chronic disorder causing persistent skin inflammation and scaling; may affect part of or the entire body; occurs in both children and adults equally.

Column	Unique Values	Count of Unique Values	Notes
erythema	[2, 3, 1, 0]	4	Ordinal scale (0–3 severity)
scaling	[2, 3, 1, 0]	4	Ordinal scale (0–3 severity)
definite_borders	[0, 3, 2, 1]	4	Ordinal scale (0–3)
itching	[3, 2, 0, 1]	4	Ordinal scale (0–3)
koebner_phenomenon	[0, 1, 2, 3]	4	Ordinal scale (0–3)
polygonal_papules	[0, 3, 2, 1]	4	Ordinal scale (0–3)
follicular_papules	[0, 3, 1, 2]	4	Ordinal scale (0–3)
oral_mucosal_involvement	[0, 3, 2, 1]	4	Ordinal scale (0–3)
knee_and_elbow_involvement	[1, 0, 3, 2]	4	Ordinal scale (0–3)
scalp_involvement	[0, 1, 2, 3]	4	Ordinal scale (0–3)
family_history	[0, 1]	2	Binary feature (0 = No, 1 = Yes)
melanin_incontinence	[0, 1, 2, 3]	4	Ordinal scale (0–3)
eosinophils_in_the_infiltrate	[0, 2, 1]	3	Ordinal scale (0–3, not all values present)
PNL_infiltrate	[0, 1, 3, 2]	4	Ordinal scale (0–3)
fibrosis_of_the_papillary_dermis	[0, 3, 1, 2]	4	Ordinal scale (0–3)
exocytosis	[3, 1, 0, 2]	4	Ordinal scale (0–3)
acanthosis	[2, 3, 1, 0]	4	Ordinal scale (0–3)
hyperkeratosis	[0, 2, 1, 3]	4	Ordinal scale (0–3)
parakeratosis	[0, 2, 3, 1]	4	Ordinal scale (0–3)
clubbing_of_the_rete_ridges	[0, 2, 1, 3]	4	Ordinal scale (0–3)
elongation_of_the_rete_ridges	[0, 2, 3, 1]	4	Ordinal scale (0–3)
thinning_of_the_suprapapillary_epidermis	[0, 2, 3, 1]	4	Ordinal scale (0–3)
spongiform_pustule	[0, 2, 1, 3]	4	Ordinal scale (0–3)
munro_microabcess	[0, 1, 2, 3]	4	Ordinal scale (0–3)
focal_hypergranulosis	[0, 2, 3, 1]	4	Ordinal scale (0–3)
disappearance_of_the_granular_layer	[0, 3, 2, 1]	4	Ordinal scale (0–3)
vacuolisation_and_damage_of_basal_layer	[0, 2, 3, 1]	4	Ordinal scale (0–3)
spongiosis	[3, 0, 2, 1]	4	Ordinal scale (0–3)
saw-tooth_appearance_of_retes	[0, 2, 3, 1]	4	Ordinal scale (0–3)
follicular_horn_plug	[0, 1, 2, 3]	4	Ordinal scale (0–3)
perifollicular_parakeratosis	[0, 2, 1, 3]	4	Ordinal scale (0–3)
inflammatory_monoluclear_inflitrate	[1, 2, 3, 0]	4	Ordinal scale (0–3)
band-like_infiltrate	[0, 3, 1, 2]	4	Ordinal scale (0–3)
Age	[0, 7, 8, 9, …, 75, NaN] (various integers/floats)	60+	Continuous numeric variable; has some missing values (NaN).
class	[1, 2, 3, 4, 5, 6]	6	Target variable: six skin disorders (multiclass classification).

	proportion
class
1	30.601093
3	19.672131
2	16.666667
5	14.207650
4	13.387978
6	5.464481

	0	1	2	3	4
erythema	2	3	2	2	2
scaling	2	3	1	2	3
definite_borders	0	3	2	2	2
itching	3	2	3	0	2
koebner_phenomenon	0	1	1	0	2
polygonal_papules	0	0	3	0	2
follicular_papules	0	0	0	0	0
oral_mucosal_involvement	0	0	3	0	2
knee_and_elbow_involvement	1	1	0	3	0
scalp_involvement	0	1	0	2	0
family_history	0	1	0	0	0
melanin_incontinence	0	0	1	0	1
eosinophils_in_the_infiltrate	0	0	0	0	0
PNL_infiltrate	0	1	0	3	0
fibrosis_of_the_papillary_dermis	0	0	0	0	0
exocytosis	3	1	1	0	1
acanthosis	2	2	2	2	2
hyperkeratosis	0	0	0	0	0
parakeratosis	0	2	2	3	0
clubbing_of_the_rete_ridges	0	2	0	2	0
elongation_of_the_rete_ridges	0	2	0	2	0
thinning_of_the_suprapapillary_epidermis	0	2	0	2	0
spongiform_pustule	0	2	0	2	0
munro_microabcess	0	1	0	0	0
focal_hypergranulosis	0	0	2	0	2
disappearance_of_the_granular_layer	0	0	0	3	2
vacuolisation_and_damage_of_basal_layer	0	0	2	0	3
spongiosis	3	0	3	0	2
saw-tooth_appearance_of_retes	0	0	2	0	3
follicular_horn_plug	0	0	0	0	0
perifollicular_parakeratosis	0	0	0	0	0
inflammatory_monoluclear_inflitrate	1	1	2	3	2
band-like_infiltrate	0	0	3	0	3
Age	55	8	26	40	45
class	2	1	3	1	3

	count	mean	std	min	25%	50%	75%	max
erythema	366.0	2.068306	0.664753	0.0	2.0	2.0	2.0	3.0
scaling	366.0	1.795082	0.701527	0.0	1.0	2.0	2.0	3.0
definite_borders	366.0	1.549180	0.907525	0.0	1.0	2.0	2.0	3.0
itching	366.0	1.366120	1.138299	0.0	0.0	1.0	2.0	3.0
koebner_phenomenon	366.0	0.633880	0.908016	0.0	0.0	0.0	1.0	3.0
polygonal_papules	366.0	0.448087	0.957327	0.0	0.0	0.0	0.0	3.0
follicular_papules	366.0	0.166667	0.570588	0.0	0.0	0.0	0.0	3.0
oral_mucosal_involvement	366.0	0.377049	0.834147	0.0	0.0	0.0	0.0	3.0
knee_and_elbow_involvement	366.0	0.614754	0.982979	0.0	0.0	0.0	1.0	3.0
scalp_involvement	366.0	0.519126	0.905639	0.0	0.0	0.0	1.0	3.0
family_history	366.0	0.125683	0.331946	0.0	0.0	0.0	0.0	1.0
melanin_incontinence	366.0	0.404372	0.869818	0.0	0.0	0.0	0.0	3.0
eosinophils_in_the_infiltrate	366.0	0.139344	0.411790	0.0	0.0	0.0	0.0	2.0
PNL_infiltrate	366.0	0.546448	0.815451	0.0	0.0	0.0	1.0	3.0
fibrosis_of_the_papillary_dermis	366.0	0.336066	0.853139	0.0	0.0	0.0	0.0	3.0
exocytosis	366.0	1.368852	1.104418	0.0	0.0	2.0	2.0	3.0
acanthosis	366.0	1.956284	0.712512	0.0	2.0	2.0	2.0	3.0
hyperkeratosis	366.0	0.527322	0.757116	0.0	0.0	0.0	1.0	3.0
parakeratosis	366.0	1.289617	0.917562	0.0	1.0	1.0	2.0	3.0
clubbing_of_the_rete_ridges	366.0	0.663934	1.056829	0.0	0.0	0.0	2.0	3.0
elongation_of_the_rete_ridges	366.0	0.991803	1.162161	0.0	0.0	0.0	2.0	3.0
thinning_of_the_suprapapillary_epidermis	366.0	0.633880	1.034924	0.0	0.0	0.0	1.0	3.0
spongiform_pustule	366.0	0.295082	0.670578	0.0	0.0	0.0	0.0	3.0
munro_microabcess	366.0	0.363388	0.759721	0.0	0.0	0.0	0.0	3.0
focal_hypergranulosis	366.0	0.393443	0.849406	0.0	0.0	0.0	0.0	3.0
disappearance_of_the_granular_layer	366.0	0.464481	0.864899	0.0	0.0	0.0	1.0	3.0
vacuolisation_and_damage_of_basal_layer	366.0	0.456284	0.954873	0.0	0.0	0.0	0.0	3.0
spongiosis	366.0	0.953552	1.130172	0.0	0.0	0.0	2.0	3.0
saw-tooth_appearance_of_retes	366.0	0.453552	0.954744	0.0	0.0	0.0	0.0	3.0
follicular_horn_plug	366.0	0.103825	0.450433	0.0	0.0	0.0	0.0	3.0
perifollicular_parakeratosis	366.0	0.114754	0.488723	0.0	0.0	0.0	0.0	3.0
inflammatory_monoluclear_inflitrate	366.0	1.866120	0.726108	0.0	1.0	2.0	2.0	3.0
band-like_infiltrate	366.0	0.554645	1.105908	0.0	0.0	0.0	0.0	3.0
class	366.0	2.803279	1.597803	1.0	1.0	3.0	4.0	6.0

Feature 1	Feature 2	Correlation (r)
thinning_of_the_suprapapillary_epidermis	clubbing_of_the_rete_ridges	0.88
follicular_papules	follicular_horn_plug	0.79
perifollicular_parakeratosis	follicular_horn_plug	0.93
perifollicular_parakeratosis	follicular_papules	0.84
knee_and_elbow_involvement	clubbing_of_the_rete_ridges	0.72

Feature 1	Feature 2	Correlation (r)
thinning_of_the_suprapapillary_epidermis	class	-0.68
clubbing_of_the_rete_ridges	class	-0.67
PNL_infiltrate	class	-0.55
scalp_involvement	class	-0.53
munro_microabscess	class	-0.52

Feature 1	Feature 2	Correlation (r)
thinning_of_the_suprapapillary_epidermis	scalp_involvement	0.75
thinning_of_the_suprapapillary_epidermis	munro_microabscess	0.69
thinning_of_the_suprapapillary_epidermis	disappearance_of_the_granular_layer	0.55
thinning_of_the_suprapapillary_epidermis	parakeratosis	0.52
thinning_of_the_suprapapillary_epidermis	knee_and_elbow_involvement	0.64
clubbing_of_the_rete_ridges	scalp_involvement	0.77
clubbing_of_the_rete_ridges	munro_microabscess	0.67
clubbing_of_the_rete_ridges	disappearance_of_the_granular_layer	0.59
clubbing_of_the_rete_ridges	parakeratosis	0.54
clubbing_of_the_rete_ridges	knee_and_elbow_involvement	0.72
scaling	definite_borders	0.36
spongiform_pustule	follicular_papules	0.50
spongiform_pustule	scalp_involvement	0.46
knee_and_elbow_involvement	scalp_involvement	0.66

VIF Value	Interpretation
1	No multicollinearity
1–5	Moderate multicollinearity (generally okay)
> 5	High multicollinearity (needs investigation)
> 10	Severe multicollinearity (consider removal)

Variable	VIF
vacuolisation_and_damage_of_basal_layer	17.234025
band-like_infiltrate	13.527991
melanin_incontinence	11.344378
perifollicular_parakeratosis	11.064677
saw-tooth_appearance_of_retes	10.664021

Variable	VIF
polygonal_papules	8.516422
focal_hypergranulosis	8.221431
clubbing_of_the_rete_ridges	7.982584
follicular_horn_plug	7.918923
oral_mucosal_involvement	7.046405
thinning_of_the_suprapapillary_epidermis	6.815498
elongation_of_the_rete_ridges	6.136761

Variable	VIF
follicular_papules	4.109109
fibrosis_of_the_papillary_dermis	3.393069
knee_and_elbow_involvement	3.352390
scalp_involvement	3.204027

Variable	VIF
exocytosis	2.904057
munro_microabcess	2.526066
spongiosis	2.464172
definite_borders	2.059011
spongiform_pustule	1.934889
PNL_infiltrate	1.830142
disappearance_of_the_granular_layer	1.825448
parakeratosis	1.768826
scaling	1.711604
itching	1.644960
koebner_phenomenon	1.610212
acanthosis	1.492661
erythema	1.470500
family_history	1.301650
hyperkeratosis	1.298062
Age	1.253764
inflammatory_monoluclear_inflitrate	1.250822
eosinophils_in_the_infiltrate	1.198826

Model	Test Accuracy	Overfit Value	F1-Score (Macro)	Training Time	Observation	Recommendation
Logistic Regression	0.8919	0.0636	0.8634	0.335 sec	High test accuracy, lowest overfitting, balanced F1, very fast	Best choice for multiclass problem
Random Forest	0.8378	0.1176	0.8085	680.123 sec	Test accuracy low, overfitting increased after tuning, very slow	Not recommended
XGBoost	0.8514	0.1350	0.7974	37.716 sec	High overfitting, test accuracy drops after tuning	Not recommended
LightGBM	0.8514	0.1486	0.8088	84.086 sec	Overfits heavily, test accuracy drops after tuning	Not recommended

S.No	Feature Name	Reason for Choosing
1	exocytosis	Indicates migration of inflammatory cells into the epidermis, a key histopathological marker.
2	munro_microabcess	Represents neutrophilic infiltration in the epidermis, characteristic of many skin disorders.
3	spongiosis	Shows intercellular edema in the epidermis, important for diagnosing inflammatory skin conditions.
4	definite_borders	Clinical feature that helps differentiate types of lesions in skin disorders.
5	spongiform_pustule	Presence of pustules is a distinctive marker of certain skin conditions.
6	PNL_infiltrate	Reflects polymorphonuclear leukocyte infiltration, important for immune response assessment.
7	disappearance_of_the_granular_layer	Histopathological change often observed in chronic skin disorders.
8	parakeratosis	Indicates abnormal keratinization, a key diagnostic feature in many skin diseases.
9	scaling	Symptom reflecting epidermal turnover and skin dryness; clinically relevant.
10	itching	Common patient-reported symptom that impacts disease severity assessment.
11	koebner_phenomenon	Clinical phenomenon where lesions appear at sites of trauma, specific to certain disorders.
12	acanthosis	Thickening of the epidermis, a histological marker of chronic skin changes.
13	erythema	Redness of the skin indicating inflammation; important clinical sign.
14	family_history	Genetic predisposition can influence susceptibility to skin disorders.
15	hyperkeratosis	Thickening of the stratum corneum, a common histopathological feature.
16	Age	Age can affect disease prevalence and presentation; useful demographic predictor.
17	inflammatory_monoluclear_inflitrate	Indicates chronic inflammation and immune system involvement.
18	eosinophils_in_the_infiltrate	Presence of eosinophils reflects immune response and disease activity in certain skin disorders.

	count
Age
35	22
40	17
50	17
27	16
36	16
22	15
55	14
25	14
30	13
33	12
60	11
52	11
42	10
18	9
34	8
20	8
8	7
10	7
51	7
45	7
62	7
32	6
47	6
19	6
46	6
17	5
44	5
48	5
56	5
28	5
16	5
41	4
7	4
70	4
43	4
38	3
21	3
26	3
29	3
23	3
12	3
15	2
57	2
65	2
9	2
39	2
13	2
31	2
61	2
37	2
24	2
53	2
64	1
0	1
67	1
58	1
49	1
63	1
68	1
75	1

	Training Accuracy	Test Accuracy	Overfit (Train - Test Acc)	Precision	Recall	F1-Score	Training Time (sec)
Logistic Regression	0.9555	0.8919	0.0636	0.8696	0.8622	0.8634	0.335
Random Forest	0.9144	0.8378	0.0765	0.8183	0.8261	0.8073	0.560
XGBoost	0.9966	0.8649	0.1317	0.8851	0.7972	0.8228	0.915
LightGBM	1.0000	0.8784	0.1216	0.8796	0.8167	0.8309	0.767

	Training Accuracy	Test Accuracy	Overfit (Train - Test Acc)	Precision	Recall	F1-Score	Training Time (sec)
Random Forest Classifier	0.9555	0.8378	0.1176	0.8190	0.8122	0.8085	680.123
XGBoost Classifier	0.9863	0.8514	0.1350	0.8207	0.7833	0.7974	37.716
LightGBM Classifier	1.0000	0.8514	0.1486	0.8108	0.8139	0.8088	84.086

Class	Precision	Recall	F1-Score	Support
1	1.00	1.00	1.00	23
2	0.77	0.83	0.80	12
3	0.82	0.93	0.88	15
4	0.88	0.70	0.78	10
5	1.00	0.90	0.95	10
6	0.75	0.75	0.75	4

PRCP-1027: Skin Disorder Prediction¶

Detecting The Type Of Skin Disoder¶

Project Type - Classification¶

Project Summary:¶

Problem Statement:¶

Let's Begin!¶

1. Know Your Data¶

1.1. Import Libraries:¶

1.2. Data Collection / Loading:¶

1.3. Basic Overview:¶

1.4. Dataset Information:¶

1.5. Domain Analysis:¶

Domain Analysis Report:¶

Clinical Attributes (12):¶

Histopathological Attributes (22):¶

Class of Diseases (1):¶

2. Data wrangling / Cleaning¶

2.1. Change the dtype:¶

2.2. Extracting categorical and numerical columns¶

2.3. Observations from Numeric Columns and Imputation¶

2.4. Check for and remove duplicate values¶

3. Task 1 - Exploratory Data Analysis (EDA)¶

3.1. Univariate Analysis: Investigating Individual Features¶

3.1.1. Clinical Features¶

Chart-1. Distribution of Clinical Features¶

Observations:¶

3.1.2. Age Feature¶

Chart-2. Distribution of Age Feature¶

Observations:¶

3.1.3. Histopathological Features¶

Chart-3. Distribution of Histopathological Features¶

Observation:¶

3.1.4. Barplot For Class(Target) Feature¶

Chart-4. Distribution of Class Feature¶

Observations:¶

3.1.5. PiePlot For Class(Target) Feature¶

Chart-5. Pie Chart Distribution of Class Feature¶

3.1.6. Distribution For Age Feature¶

Chart-6. Visualize the distribution of Age feature¶

Observations:¶

3.2. Bivariate Analysis: Examining Relationships Between Variable Pairs¶

3.2.1. Regression plot of Age feature vs Target Variable¶

Chart-7. Regression plot of Age feature vs Target Variable¶

3.2.2. All Features vs Target Variable¶

Chart-8. All Features vs Target Variable¶

Observations:¶

3.2.3. Age Feature vs Target Variable¶

Chart-9. Age Feature vs Target Variable¶

Observations:¶

3.3. Multivariate Analysis: Examines multiple variables simultaneously¶

3.3.1. Correlation Heatmap: Highlights correlations between numerical features¶

Chart-10. Correlation Heatmap¶

3.4. Hypothesis Testing¶

Based on the chart experiments, define three hypothetical statements about the dataset. In the next three answers, perform hypothesis testing to obtain a final conclusion about the statements through your code and statistical testing.¶

3.4.1. Hypothetical Statement - 1¶

1. State Your research hypothesis as a null hypothesis and alternate hypothesis.¶

2. Perform an appropriate statistical test¶

3. Business Insight:¶

3.4.2. Hypothetical Statement - 2¶

1. State Your research hypothesis as a null hypothesis and alternate hypothesis.¶

2. Perform an appropriate statistical test¶

3. Business Insight:¶

3.4.3. Hypothetical Statement - 3¶

1. State Your research hypothesis as a null hypothesis and alternate hypothesis.¶

2. Perform an appropriate statistical test¶

3. Business Insight:¶

4. Data Pre-Processing¶

4.1. Handling Missing Values / Null Values¶

4.2. Handling Outliers: Detection and Treatment Strategies¶

4.2.1. Perform outlier detection:¶

Chart-11. Plotting box plots for all numerical variable¶

4.2.2. Calculate the number of outliers and their percentage:¶

5. Feature Engineering¶

5.1. Check if my target feature is imbalanced or not¶

Chart-13. Target Variable Distribution (Class)¶

5.2. Feature Selection¶

5.2.1. Encoding Categorical and Numerical Variables¶

5.2.2. Correlation Heatmap of Features¶

Chart-14. Correlation Heatmap of Features¶

Insights from Heatmap¶

	count
Age
35	22
40	17
50	17
27	16
36	16
22	15
55	14
25	14
30	13
33	12
60	11
52	11
42	10
18	9
34	8
20	8
8	7
10	7
51	7
45	7
62	7
32	6
47	6
19	6
46	6
17	5
44	5
48	5
56	5
28	5
16	5
41	4
7	4
70	4
43	4
38	3
21	3
26	3
29	3
23	3
12	3
15	2
57	2
65	2
9	2
39	2
13	2
31	2
61	2
37	2
24	2
53	2
64	1
0	1
67	1
58	1
49	1
63	1
68	1
75	1